An introduction to the new Results Table integrated into the output cell of Notebooks, powered by the VS Code extension called Data Wrangler. The post Announcing the Data Wrangler powered Notebook Results Table appeared first on Microsoft for Python Developers Blog.  ( 23 min )

GitHub Spark and GitHub Copilot are powerful development tools that can significantly boost productivity even when used out of the box. However, in enterprise settings, a common request is for development support that aligns with specific compliances or regulations. While GitHub Copilot allows you to choose models like GPT-4o or others, it does not currently support the use of custom fine-tuned models. Additionally, many users might find it unclear how to integrate Copilot with external services, which can be a source of frustration. To address such needs, one possible approach is to build a custom MCP server and connect it to GitHub Copilot. For a basic “Hello World” style guide on how to set this up, please refer to the articles below. https://devblogs.microsoft.com/dotnet/build-a-model…  ( 34 min )

We’ve made a major update to User Data Functions! This update addresses the most critical feedback we have received since the start of the preview: the testing and development experience of your functions. Important note: In order to use this feature, you need to upgrade to the latest version of the fabric-user-data-functions library. Beginning in … Continue reading “Test and validate your functions with Develop mode in Fabric User Data Functions (Preview)”  ( 7 min )

The Open API specification, formerly Swagger Specification, is a widely used, language-agnostic description format for REST APIs. This allows humans and computers alike to discover and understand the capabilities of a service in a standardized format. This is critical for creating integrations with external systems, AI agents and code generators. Now, you can automatically generate … Continue reading “OpenAPI specification code generation now available in Fabric User Data Functions”  ( 6 min )

If you’ve already built Model Context Protocol (MCP) servers with the MCP SDKs and wished you could turn them into world class Remote MCP servers using a hyperscale, serverless platform, then this one’s for you! We’ve published samples showing how to host bring‑your-own (BYO) Remote MCP servers on Azure Functions, so you can run the servers you’ve already built with the MCP SDKs—Python, Node, and .NET—with minimal changes and full serverless goodness. Why this is exciting Keep your code. If you’ve already implemented servers with the MCP SDKs (Python, Node, .NET), deploy them to Azure Functions as remote MCP servers with just one line of code change. Serverless scale when you need it. Functions on the Flex Consumption plan handles bursty traffic, scales out and back to zero automatically,…  ( 22 min )

What is Foundry Local? Foundry Local is a user-friendly tool that helps you run small AI language models directly on your Windows or Mac computer. Think of it as a way to have your own personal ChatGPT running locally on your machine, without needing an internet connection or sending your data to external servers. Currently, Foundry Local works great with several popular model families: Phi models (Microsoft's small but powerful models) Qwen models (Alibaba's multilingual models) DeepSeek models (Efficient reasoning models) In this tutorial, we'll learn how to set up the Qwen3-0.6B model step by step. Don't worry if you're new to AI - we'll explain everything along the way! Why Do We Need to Convert AI Models? When you download AI models from websites like Hugging Face (think of it as G…  ( 37 min )

Introduction The integration of AI into web applications has reached a new milestone with the introduction of Model Context Protocol (MCP) support in Azure AI Foundry Agent Service. This powerful new feature allows developers to extend the capabilities of their Azure AI Foundry agents by connecting them to tools hosted on remote MCP servers—and the best part? Both your applications and MCP servers can be hosted seamlessly on Azure App Service. No custom code is required to get your agents hooked up to these MCP servers either, it's built right into this new functionality. In this blog post, we'll explore how this new functionality works, demonstrate a practical implementation, and show you how to get started with your own MCP-enabled applications on App Service. What is Model Context Proto…  ( 32 min )

Function calling is a powerful technique that allows Large Language Models (LLMs) to go beyond their static training data and interact with real-world, dynamic information sources like databases and APIs. This capability turns a simple chat interface into a powerful tool for fetching real-time data, executing code, and much more. The process of Tool/Function calling typically involves two main components: a client application and the LLM. A user's request, such as "Do I need to carry an umbrella in Bangalore today?", is sent from the client application to the LLM. Along with this message, a tool definition is provided. This definition gives the LLM the context it needs to understand what tools are available along usage methods. The LLM analyses the user's request and the list of available …  ( 34 min )

Special ThanksThanks to Andrew Thomas, Kurt Niebuhr, and Sachi Desai for their invaluable support, insightful discussions, and for providing the compute resources that made testing this deployment possible. Your contributions were essential in bringing this project to life.Introduction OpenAI recently released GPT-OSS, its first open-source large language model. With the rise of high-performance GPUs in the cloud, running advanced AI inference workloads has become easier than ever. Microsoft Azure’s AKS (Azure Kubernetes Service) paired with KAITO (Kubernetes AI Toolchain Operator) provides a powerful, scalable environment for deploying such models. KAITO simplifies provisioning GPU nodes, managing inference workloads, and integrating AI-optimized runtimes like vLLM. In this tutorial, we’l…  ( 28 min )

This blog covers a niche issue with Next.js, but potentially any node.js-based app, when using the Node.js URI API  ( 6 min )

The original post (Japanese) was written on 13 August 2025. Zero code instrumentationでMicronautアプリケーションからAzure Monitorにtraceやmetricを送信したい – Logico Inside This entry is a series posts below. Please take a look for background information. Send signals from Micronaut native image applications to Azure Monitor | Microsoft Community Hub Received another question from the customer. I understand that I can get metrics and traces, but is it possible to send them to Azure Monitor (Application Insights) without using code? If you are not familiar with zero-code instrumentation, please check the following URL. Zero-code | OpenTelemetry The customer wondered if the dependencies would take care of everything else when they only specified the dependencies and destinations. To confirm this (and to prov…  ( 36 min )

The original post (Japanese) was written on 29 July 2025. MicronautからAzure Monitorにlogを送信したい – Logico Inside This entry is related to the following one. Please take a look for background information. Send signals from Micronaut native image applications to Azure Monitor | Microsoft Community Hub Where can we post logs? Log destination differs depending upon managed services such as App Service, Container Apps, etc. We can also send logs to specified destination which is different destination from the default one. In the case of Azure Container Apps, for instance, we have several options to send logs. TypeDestinationHow to Write console output to a log ContainerAppConsoleLogs_CL If diagnostic settings are configured, destination table may differ from the above. The output destination can …  ( 40 min )

The original post (Japanese) was written on 23 July 2025. MicronautからAzure Monitorにtraceを送信したい – Logico Inside This entry is related to the following one. Please take a look for background information. Send signals from Micronaut native image applications to Azure Monitor | Microsoft Community Hub Prerequisites Maven: 3.9.10 JDK version 21 Micronaut: 4.9.0 or later The following tutorial was used as a reference. OpenTelemetry Tracing with Oracle Cloud and the Micronaut Framework As of 13 August 2025, GDK (Graal Dev Kit) guides are also available. Create and Trace a Micronaut Application Using Azure Monitor Create an archetype We can create an archetype using Micronaut’s CLI (mn) or Micronaut Launch. In this entry, use application.yml instead of application.properties for application conf…  ( 31 min )

The original post (Japanese) was written on 20 July 2025. MicronautアプリケーションからAzure Monitorにlog/metric/traceを送信したい – Logico Inside Query One of my customers, using Micronaut to develop their applications, asked me about the following question. We are currently creating microservices with Micronaut and orchestrating them on Azure (Azure Container Apps). We would like to ask about the following two points. We found information on how to send logs and metrics to Azure Monitor in the Micronaut documentation and guides, but we need to know how to send traces. Is it also available with GraalVM Native Image? Their query is, in short, Spring Boot provides Spring Boot Starter and Quarkus provides Quarkus as dependencies, but does Micronaut do the same? Spring Boot starter Quarkus Opentelemetry E…  ( 24 min )

The original post (Japanese) was written on 20 July 2025.MicronautからAzure Monitorにmetricを送信したい – Logico Inside This entry is related to the following one. Please take a look for background information.Send signals from Micronaut native image applications to Azure Monitor | Microsoft Community Hub Prerequisites Maven: 3.9.10 JDK version 21 Micronaut: 4.9.0 or later The following tutorials were used as a reference. Create a Micronaut Application to Collect Metrics and Monitor Them on Azure Monitor Metrics Collect Metrics with Micronaut Create an archetype We can create an archetype using Micronaut’s CLI (mn) or Micronaut Launch. In this entry, use application.yml instead of application.properties for application configuration. So, we need to specify the feature “yaml” so that we can includ…  ( 37 min )

1. Weekly Highlights This episode kicked off with the top news and updates in the Azure AI ecosystem: GPT-5 and GPT-OSS Models Now in Azure AI Foundry: Azure AI Foundry now supports OpenAI’s GPT-5 lineup (including GPT-5, GPT-5 Mini, and GPT-5 Nano) and the new open-weight GPT-OSS models (120B, 20B). These models offer powerful reasoning, real-time agent tasks, and ultra-low latency Q&A, all with massive context windows and flexible deployment via the Model Router. Flux 1 Context Pro & Flux 1.1 Pro from Black Forest Labs: These new vision models enable in-context image generation, editing, and style transfer, now available in the Image Playground in Azure AI Foundry. Browser Automation Tool (Preview): Agents can now perform real web tasks—search, navigation, form filling, and more—via na…  ( 32 min )

We’re excited to share that we’ve entered a new chapter in how we bring you Azure AI content! To streamline your experience, several Azure AI Tech Community blogs have been consolidated into a single destination—the Azure AI Foundry Blog. What’s Changed Multiple Azure AI blogs have merged into one. The Azure AI technical blogs – including the AI Platform Blog, Azure AI Services Blog, and Azure Machine Learning Blog – are now part of the new Azure AI Foundry Blog. By consolidating these resources into a single blog, you have one place to find technical news, tutorials, and announcements. Seamless Access to Past Content All existing content has been preserved. Posts from the legacy blogs have been migrated into the Azure AI Foundry Blog archive. Older posts remain accessible through redirect…  ( 21 min )

We’re excited to share that we’ve entered a new chapter in how we bring you Azure AI content! To streamline your experience, several Azure AI Tech Community blogs have been consolidated into a single destination—the Azure AI Foundry Blog. What’s Changed Multiple Azure AI blogs have merged into one. The Azure AI technical blogs – including the AI Platform Blog, Azure AI Services Blog, and Azure Machine Learning Blog – are now part of the new Azure AI Foundry Blog. By consolidating these resources into a single blog, you have one place to find technical news, tutorials, and announcements. Seamless Access to Past Content All existing content has been preserved. Posts from the legacy blogs have been migrated into the Azure AI Foundry Blog archive. Older posts remain accessible through redirect…  ( 21 min )

Building on our previous post about implementing dev-to-prod promotion with GitHub Actions, this follow-up demonstrates the same “build once, deploy everywhere” pattern using Azure DevOps Pipelines. You’ll learn how to leverage Azure DevOps YAML pipelines with Azure Developer CLI (azd). This approach ensures consistent, reliable deployments across environments. Environment-Specific Infrastructure The infrastructure approach is identical […] The post Azure Developer CLI: From Dev to Prod with Azure DevOps Pipelines appeared first on Azure DevOps Blog.  ( 25 min )

We’re making an important change to how Azure DevOps displays OAuth client secrets to align with industry best practices and improve our overall security posture. Starting September, newly generated client secrets will be shown only once at the time of creation. After that, they will no longer be retrievable via the UI or API. This […] The post Azure DevOps OAuth Client Secrets Now Shown Only Once appeared first on Azure DevOps Blog.  ( 21 min )

If you’ve ever waded through a swamp of secret scanning alerts wondering, “Which of these are actually dangerous right now?” — this enhancement is for you. Secret validity checks in GitHub Advanced Security for Azure DevOps (and the standalone Secret Protection experience) add a high‑signal field to each alert: Active (still usable), or Unknown (couldn’t […] The post Hunting Living Secrets: Secret Validity Checks Arrive in GitHub Advanced Security for Azure DevOps appeared first on Azure DevOps Blog.  ( 24 min )

OpenAI recently introduced gpt-oss as an open-weight language model that delivers strong real-world performance at low cost. Available under the flexible Apache 2.0 license, these models outperform similarly sized open models on reasoning tasks, demonstrate strong tool use capabilities, and are optimized for efficient deployment on consumer hardware; see the announcement: https://openai.com/index/introducing-gpt-oss/. It’s an excellent choice for scenarios where you want the security and efficiency of a smaller model running on your application instance — while still getting impressive reasoning capabilities. By hosting it on Azure App Service, you can take advantage of enterprise-grade features without worrying about managing infrastructure: Built-in autoscaling Integration with VNet Ent…  ( 26 min )

Azure App Testing centralizes functional and performance testing for your apps. With Playwright Workspaces, you can author, execute, and analyze browser tests at scale, both locally and in the cloud, with shared reporting and parallel, cloud-hosted browsers. If you’re new to the service, start with this Overview and QuickStart. Create a Playwright Workspace (portal quick tour) In the Azure portal, search for “Azure App Testing”, open it, and select “Create.” Choose “Playwright Workspaces” (use “Azure Load Testing” for performance/load). Provide a name, region, and subscription, then create the workspace. Open your Playwright Workspace to author and run tests locally or on cloud-hosted browsers, and view results in “Test runs.” Note: A workspace lets you publish test results and artifact…  ( 27 min )

Copy Job has become the go-to solution in Microsoft Fabric Data Factory for simplified data movement, whether you’re moving data across clouds, from on-premises systems, or between services. With native support for multiple delivery styles, including bulk copy, incremental copy, and change data capture (CDC) replication, Copy Job offers the flexibility to handle a wide … Continue reading “Simplifying Data Ingestion with Copy job – Reset Incremental Copy, Auto Table Creation, and JSON Format Support”  ( 7 min )

Most developers today face a common challenge when integrating AI into their applications: the gap between natural language prompts and actual code execution. While services like OpenAI's ChatGPT excel at generating responses, they can't directly interact with your existing systems, databases, or business logic. You're left building complex orchestration layers, managing function calls manually, and creating brittle workflows that break when requirements change. Microsoft Semantic Kernel changes this paradigm entirely. Unlike traditional LLM integrations where you send a prompt and receive text, Semantic Kernel acts as an AI orchestration layer that bridges natural language with your existing codebase. Semantic Kernel intelligently decides which of your trusted functions to execute, chains…  ( 36 min )

We’re thrilled to announce that Continuous Access Evaluation (CAE) is now supported on Azure DevOps, bringing a new level of near real-time security enforcement to your development workflows. 🔐 What Is CAE? Continuous Access Evaluation (CAE) is a feature from Microsoft Entra ID that enables near real-time enforcement of Conditional Access policies. Traditionally, Microsoft Entra […] The post Real-Time Security with Continuous Access Evaluation (CAE) comes to Azure DevOps appeared first on Azure DevOps Blog.  ( 23 min )

When deploying AKS clusters, a common concern is the amount of IP address space required. If you are deploying your AKS cluster into your corporate network, the size of the IP address space you can obtain may be quite small, which can cause problems with the number of pods you are able to deploy. The simplest and most common solution to this is to use an overlay network, which is fully supported in AKS. In an overlay network, pods are deployed to a private, non-routed address space that can be as large as you want. Translation between the routable and non-routed network is handled by AKS. For most people, this is the best option for dealing with IP addressing in AKS, and there is no need to complicate things further. However, there are some limitations with overlay networking, primarily th…  ( 39 min )

This post will cover why you may see the message of “The name ‘ContainerAppConsoleLogs_CL’ does not refer to any known table, tabular variable or function.”  ( 3 min )

AzCopy is a powerful and performant tool for copying data between Azure Storage and Microsoft OneLake, and is the preferred tool for large-scale data movement due to its ease of use and built-in performance optimizations. AzCopy now supports copying data from firewall-enabled Azure Storage accounts into OneLake using trusted workspace access. Now you can use … Continue reading “Load data from network-protected Azure Storage accounts to Microsoft OneLake with AzCopy”  ( 6 min )

We’re thrilled to share a major update and simplification to OneLake’s capacity utilization model that will make it even easier to manage Fabric capacity and scale your data workloads. We are reducing the consumption rate of OneLake transactions via proxy to match the rate for transactions via redirect. This means you no longer have to … Continue reading “OneLake costs simplified: lowering capacity utilization when accessing OneLake”  ( 7 min )

Reviewed by Imran Masud and Sumit Sarabhai When it comes to working with Microsoft SQL Server in Python, pyodbc has long been the de facto driver. It’s mature, trusted and has been serving the community well for years. But as applications scale and Python becomes more central to modern data workflows — from microservices to […] The post mssql-python vs pyodbc: Benchmarking SQL Server Performance appeared first on Microsoft for Python Developers Blog.  ( 25 min )

Data is at the heart of every application. But evolving your schema is risky business. One broken migration, and your dev or prod environment can go down. We've all experienced it: mismatched columns, orphaned constraints, missing fields, or that dreaded "table already exists" error. But what if debugging migrations didn’t have to be painful? What if you could simply describe the error or broken state, and AI could fix your migration in seconds? In this blog, you’ll learn how to: Use GitHub Copilot to describe and fix broken migrations with natural language Catch schema issues like incorrect foreign keys before they block your workflow Validate and deploy your database changes using GibsonAI CLI Broken migrations are nothing new. Whether you're working on a side project or part of a larg…  ( 27 min )

OpenAI has released the open-source models gpt-oss-20b and gpt-oss-120B. Enterprises and developers can now deploy these models on edge devices without relying on cloud APIs, enabling local deployment. The AI Toolkit for VS Code extension provides developers with a complete development workflow—from model testing and local deployment to building intelligent agent applications—creating a comprehensive AI application development workflow. We used the AI Toolkit in conjunction with gpt-oss-20b to build local AI applications. Understanding gpt-oss OpenAI has released gpt-oss-120b and gpt-oss-20b, the first open-weight language models since GPT-2. Both models utilize mixture-of-experts (MoE) architecture with MXFP4 quantization, delivering exceptional reasoning capabilities and tool use functio…  ( 26 min )

We’re excited to share that customer-managed keys (CMK) for Microsoft Fabric workspaces are now available in public preview in all public regions! This expansion makes it easier for customers worldwide to meet compliance requirements and implement robust data protection strategies. Note: This feature was released in public preview in a limited set of regions earlier … Continue reading “Customer-managed keys for Fabric workspaces is now in Public Preview”  ( 5 min )

Co-author: Meenal Srivastva Managing secure, seamless access to data sources is a top priority for organizations using Microsoft Fabric. With workspace identity authentication, teams can simplify credential management, enhance security, and streamline data access across their enterprise. Workspace identity in Fabric is an automatically managed service principal associated with workspaces (excluding My Workspaces). When you … Continue reading “Introducing support for Workspace Identity Authentication in Fabric Connectors”  ( 7 min )

Co-authored with Kevin Liu, Apache Iceberg Committer and Principal Engineer at Microsoft.  Microsoft Fabric is a unified SaaS data and analytics platform designed for the era of AI. All workloads in Microsoft Fabric use Delta Lake as the standard, open-source table format. With Microsoft OneLake, Fabric’s unified SaaS data lake, customers can unify their data … Continue reading “How Microsoft OneLake seamlessly provides Apache Iceberg support for all Fabric Data”  ( 7 min )

Why Page Numbers Matter in AI Search When users search for content within large PDFs—such as contracts, manuals, or reports—they often need to know not just what was found, but where it was found. Associating search results with page numbers enables: Contextual navigation within documents. Precise citations in knowledge bases or chatbots. Improved user trust in AI-generated responses. Prerequisites for Azure Blob Storage & Azure AI Search Setup Summary 1. Azure Blob Storage A container is configured to store PDF files. 2. Appropriate permissions: The AI search service must have Storage Blob Data Reader access to the container. If using RBAC, ensure the managed identity is properly assigned.  Ref:  AI search Search-blob-indexer-role-based-access How to Index Azure Blobs   Technical Ap…  ( 27 min )

Background Troubleshooting outbound connectivity issues in Azure Kubernetes Service (AKS) can be daunting. The complexity arises from the layered nature of Azure networking, which can span NSGs, UDRs, route tables, firewalls, and private endpoints. When a pod fails to reach an external service, pinpointing the root cause often requires deep familiarity with Azure networking and manual inspection of multiple network resources. This complex process slows down resolution and increases operational overhead for cluster operations and platform teams. Announcing the Connectivity Analysis feature for AKS To simplify the network troubleshooting process, we're excited to announce the Connectivity Analysis feature for AKS, now available in Public Preview. This feature leverages the same underlying en…  ( 27 min )

The August 2025 release includes Python shell integration support for Python 3.13+, Python Environments extension improvements, enhanced terminal suggestions with documentation, and more! The post Python in Visual Studio Code – August 2025 Release appeared first on Microsoft for Python Developers Blog.  ( 23 min )

Azure App Testing enables developers and QA teams to run large-scale functional and performance tests to pinpoint issues in their applications, across frameworks like Playwright, JMeter, or Locust. It brings together two powerful testing capabilities—Azure Load Testing and Microsoft Playwright Testing—into a single hub in the Azure Portal, providing a consistent experience for resource provisioning, access control, and consolidated billing. With Azure App Testing, our goal is to help you spend less time managing infrastructure and more time harnessing AI-driven test automation to boost quality and innovation. Key Benefits AI-driven testing: Accelerate test creation and insights with AI-powered tooling. Limitless scale: Simulate real-world traffic from multiple regions with load tests, and…  ( 23 min )

Core Topics Discussed 1. Weekly Highlights RFD Observability tools in Azure AI Foundry: Real-time model telemetry, auto evals, quick evals, Python grader. GitHub Copilot Pro with Spark: AI pair programmer for code explanation and workflow suggestions. Synthetic Data for Vision Models: Training accurate models with procedurally generated data. Agent-Friendly Websites: Making sites accessible to AI agents via APIs, semantic markup, and OpenAPI specs. MCP (Model Context Protocol): Standardizing agent memory and context for scalable AI. 2. Spotlight: Foundry Local – On-Device AI What is Foundry Local?A toolkit to run open-source AI models directly on your device (CPU, GPU, NPU) for fast, private, and efficient inference. Why does it matter?Developers can work with data locally, switch seaml…  ( 29 min )

If you have been following this blog post series then you should have a working Terraform config from the first two posts, plus a managed identity from the third post that has the correct authorizations as a workload identity for use in deployment pipelines. In this last post we will configure OpenID Connect for the … Continue reading “Terraform Provider for Microsoft Fabric: #4 Deploying a Fabric config with Terraform in GitHub Actions”  ( 9 min )

Introducing Dev Proxy v1.0, with new language model-specific testing capabilities to help developers build more reliable AI-powered applications by simulating real-world scenarios and tracking resource usage. The post Dev Proxy v1.0 with new features for building robust AI-powered apps appeared first on Microsoft 365 Developer Blog.  ( 26 min )

Prerequisite This article assumes that you’ve already gone through a following post. Please make sure to read it before proceeding: Developing a Custom Engine Agent for Microsoft 365 Copilot Chat Using Pro-Code With the article, we have found how to publish a Custom Engine Agent using pro-code approaches such as C#. In this post, I’d like to shift the focus to security, specifically how to protect the endpoint of our custom Microsoft 365 Copilot. Through several architectural explorations, we found an approach that seems to work well. However, I strongly encourage you to review and evaluate it carefully for your production environment. Which Endpoints Can Be Controlled? In the current architecture, there are three key endpoints to consider from a security perspective: Teams EndpointThis…  ( 53 min )

Prerequisite This article assumes that you’ve already gone through a following post. Please make sure to read it before proceeding: Developing a Custom Engine Agent for Microsoft 365 Copilot Chat Using Pro-Code With the article, we have found how to publish a Custom Engine Agent using pro-code approaches such as C#. In this post, I’d like to shift the focus to security, specifically how to protect the endpoint of our custom Microsoft 365 Copilot. Through several architectural explorations, we found an approach that seems to work well. However, I strongly encourage you to review and evaluate it carefully for your production environment. Which Endpoints Can Be Controlled? In the current architecture, there are three key endpoints to consider from a security perspective: Teams EndpointThis…  ( 53 min )

Just yesterday, OpenAI announced the release of gpt-oss-120b and gpt-oss-20b, two new state-of-the-art open-weight language models. These models are designed to run on lighter weight GPU resources making them highly accessible for developers who want to self-host powerful language models within their own environments. If you’re looking to deploy these models in the cloud, Azure Container Apps serverless GPUs are a great option. With support for both A100 and T4 Open AI gpt-oss-120b running on Azure Container Apps serverless GPUsGPUs, serverless GPUs support both the gpt-oss-120b and gpt-oss-20b models, providing a cost-efficient and scalable platform with minimal infrastructure overhead. Open AI gpt-oss-120b running on Azure Container Apps serverless GPUs In this blog post, we’ll walk thro…  ( 31 min )

A recent issue was identified in the azure-resourcemanager libraries for Java regarding HTTP headers exceeding 8,192 bytes. Following analysis, this was addressed within the Azure SDK for Java by increasing the maximum response header size limit of “reactor-netty-http” to 256 KB, accommodating larger headers. To benefit from this mitigation, it is necessary to upgrade azure-core-http-netty […] The post “HTTP header exceeding 8,192 bytes” error when using the Azure SDK for Java appeared first on Microsoft for Java Developers.  ( 21 min )

Wassette empowers AI agents to securely fetch and run Wasm tools, enabling dynamic, permission-controlled capabilities with zero dependencies. The post Introducing Wassette: WebAssembly-based tools for AI agents appeared first on Microsoft Open Source Blog.  ( 14 min )

LangChain is a powerful framework that simplifies the development of applications powered by large language models (LLMs). It provides essential building blocks like chains, agents, and memory components that enable developers to create sophisticated AI workflows beyond simple prompt-response interactions. LangChain's importance lies in its ability to orchestrate complex AI operations, integrate multiple data sources, and maintain conversation context—making it the go-to choice for production-ready AI applications. In this blog post, we'll explore a sample application that demonstrates how you can easily deploy a LangChain application integrated with Azure OpenAI Foundry models to Azure App Service. We'll walk through this complete example that showcases a conversational AI chat interface …  ( 28 min )

Welcome to KubeCon + CloudNativeCon India 2025! We’re thrilled to join this year’s event in Hyderabad as a Gold sponsor, where we’ll be highlighting the newest innovations in Azure and Azure Kubernetes Service (AKS) while connecting with India’s dynamic cloud-native community. We’re excited to share some powerful new AKS capabilities that bring AI innovation to the forefront, strengthen security and networking, and make it easier than ever to scale and streamline operations. Innovate with AI AI is increasingly central to modern applications and competitive innovation, and AKS is evolving to support intelligent agents more natively. The AKS Model Context Protocol (MCP) server, now in public preview, introduces a unified interface that abstracts Kubernetes and Azure APIs, allowing AI agents …  ( 28 min )

Inbound IPv6 support on public multi-tenant App Service has been in public preview for a while now, so we're excited to finally be able to announce that it is now generally available across all public Azure regions for multi-tenant apps on all Basic, Standard, and Premium SKUs, Functions Consumption, Functions Elastic Premium, and Logic Apps Standard! The limitations called out in the previous blog post have been removed except for IP-SSL IPv6 bindings still not being supported. How it works IPv6 inbound requires two things: an IPv6 address that accepts traffic coming in, and a DNS record that returns an IPv6 (AAAA) record. You’ll also need a client that can send and receive IPv6 traffic. This means that you may not be able to test it from your local machine since many networks today only …  ( 21 min )

IN ONE hour, all ONLINE, complete our new training for Generative AI for Beginners – Java Edition – part of the incredible Microsoft Generative AI series – Python, .NET and JavaScript. In this training get first-hand experience with Core Generative AI Techniques and Tools we’re using inside Microsoft: Foundry Local MCP Responsible AI Quick Start Star and […] The post Generative AI for Beginners – Java Edition launched appeared first on Microsoft for Java Developers.  ( 22 min )

LangChain is a powerful framework that simplifies the development of applications powered by large language models (LLMs). It provides essential building blocks like chains, agents, and memory components that enable developers to create sophisticated AI workflows beyond simple prompt-response interactions. LangChain’s importance lies in its ability to orchestrate complex AI operations, integrate multiple data sources, and maintain conversation context—making it the go-to choice for production-ready AI applications.  ( 5 min )

We’re excited to announce that Azure Monitor Network Security Perimeter features are now generally available! This update is an important step forward for Azure Monitor’s security, providing comprehensive network isolation for your monitoring data. In this post, we’ll explain what Network Security Perimeter is, why it matters, and how it benefits Azure Monitor users. Network Security Perimeter is purpose-built to strengthen network security and monitoring, enabling customers to establish a more secure and isolated environment. As enterprise interest grows, it’s clear that this feature will play a key role in elevating the protection of Azure PaaS resources against evolving security threats. What is Network Security Perimeter and Why Does It Matter? Network Security Perimeter is a network i…  ( 31 min )

Azure Monitor logs are trusted by hundreds of thousands of organizations to monitor mission-critical workloads. But with such a diverse customer base, there’s no one-size-fits-all solution. That’s why we’re excited to announce a series of major advancements to Auxiliary Logs, the Azure Monitor plan that is designed for high volume logs. Auxiliary Logs works in tandem with all other Azure Monitor tools including the more powerful Basic and Analytics Logs plans. Together, they are the one-stop-shop for all the logging needs of an organization. Auxiliary Logs were introduced last year and have gained a lot of traction since. There are many customers that ingest data into Auxiliary Logs, with several teams ingesting more than a petabyte of logs per day. Over the last few months, we have moved …  ( 27 min )

Batch users often encounter issues like nodes suddenly gets into unusable state due to high CPU or Disk usage. Alerts allow you to identify and address issues in your system. This blog will focus on how users can enable alerts when the node is consuming high amount of disk by configuring the threshold limit. With this user can get notified beforehand when the node gets into unusable state and pre-emptively takes measures to avoid service disruptions. The task output data is written to the file system of the Batch node. When this data reaches more than 90 percent capacity of the disk size of the node SKU, the Batch service marks the node as unusable and blocks the node from running any other tasks until the Batch service does a clean up. The Batch node agent reserves 10 percent capacity of …  ( 29 min )

Overview Semantic models are evolving to work more seamlessly with Mirrored artifacts—giving you greater flexibility, control, and transparency when working with mirrored data. Why are we decoupling semantic models from Mirroring Artifacts? Historically, Mirrored artifacts were created with an automatically ‘coupled’ semantic model. While convenient out-of-the-box, this approach limited how you could shape, interpret, and … Continue reading “Decoupling Semantic Model for Mirroring Customers”  ( 6 min )

Introduction Welcome to What’s New in Fabric Warehouse, where we’ll spotlight our work improving quality, delivering major performance enhancements, boosting developer productivity, and our continuous investments in security. Whether you’re migrating from Synapse, optimizing your workloads, writing SQL in VS Code, or exploring new APIs, this roundup has something for every data professional. With quality … Continue reading “What’s new in Fabric Warehouse – July 2025 Recap”  ( 10 min )

Microsoft Fabric’s Warehouse and SQL Endpoint now support both Case Sensitive (CS) and Case Insensitive (CI) collations, providing users with powerful tools to ensure data consistency and flexibility. This blog delves into the upcoming change – updated collation configuration requirements at the workspace and item levels, offering insights into how these features enhance data workflows, … Continue reading “Coming soon: Streamlining Data Management with Collation Settings in Microsoft Fabric Warehouse and SQL Endpoint”  ( 7 min )

Multiple Scheduler: A Game-Changer for Complex Scheduling Needs In today’s data environments, one size rarely fits all. Yet until now, Fabric only allowed one scheduler per item, forcing users to duplicate pipelines, manually configure jobs, or build brittle workarounds to meet real-world needs. That changes today. We’re excited to announce Multiple Scheduler support in Fabric — a long-awaited … Continue reading “Unlocking Flexibility in Fabric: Introducing Multiple Scheduler and CI/CD Support”  ( 6 min )

Any experiences that require additional setup is cumbersome, especially when there are multiple people needed. In GitHub Advanced Security for Azure DevOps, we’re working to make it easier to enable features and scale out enablement across your enterprise. You can now automatically inject the dependency scanning task into any pipeline run targeting your default branch. […] The post Automate your open-source dependency scanning with Advanced Security appeared first on Azure DevOps Blog.  ( 23 min )

We have officially ended the support for TLS 1.1 and earlier on the Fabric platform. As previously announced, starting July 31, 2025, all outbound connections from Fabric to customer data sources must use TLS 1.2 or later. This update follows our earlier announcement in the TLS Deprecation for Fabric blog, where we outlined the rationale and timeline for this … Continue reading “Fabric Platform Support for Transport Layer Security (TLS) 1.1 and earlier versions has ended”  ( 5 min )

Mirroring for Google BigQuery is now available in Private Preview—bringing the power of Microsoft Fabric to your BigQuery data with zero-ETL, near real-time replication, and native integration across the Fabric experience. This release marks a major milestone in our cross-cloud data strategy, enabling customers to unify analytics across various platforms without the complexity of traditional data movement. What’s New With Mirroring … Continue reading “Mirroring for Google BigQuery in Microsoft Fabric (Private Preview)”  ( 5 min )

Why MCP & Copilot for Azure Matter  What is AI-Assisted Azure Development? AI-assisted development means you can use natural language to manage Azure resources. The Azure MCP Server acts as a bridge between AI agents (like Copilot) and Azure services, exposing tools in a standard way. GitHub Copilot for Azure lets you chat with your cloud—no need to memorize commands or scripts. Key takeaway:These tools make Azure more accessible and powerful. You can ask questions, deploy resources, and get best-practice advice—all in plain English.   Technical Deep Dive: How Does Azure MCP Server Work? Implements the Model Context Protocol (MCP): Standardizes how AI agents talk to Azure services. Tool Discovery: Exposes a list of available tools (like listing VMs, deploying models, checking status). Na…  ( 31 min )

Please refer to my repo to get more AI resources, wellcome to star it: https://github.com/xinyuwei-david/david-share.git  This repository ties together the entire methodology with a minimalistic example: first identify computational hotspots on the CPU, then rewrite loops characterized by "high parallelism, sequential memory access, and simple branching" into CUDA kernels to offload CPU workload and unleash GPU computing power. Running the code once can compare CPU and GPU execution times, separate transfer and computation overheads, and verify result errors. This quickly validates the feasibility and expected gains of "CPU → GPU" migration, providing a template for subsequent large-scale migration, pipeline optimization, and MIG resource partitioning in a real business environment. Overal…  ( 57 min )

At Microsoft Build 2025, we announced Azure SRE Agent, an AI tool designed to streamline incident response, improve service uptime, and cut operational costs. The SRE Agent uses AI Models in Azure AI Foundry to quickly analyze logs and metrics for root cause analysis and issue resolution. This advanced AI agent enhances incident and infrastructure management in Azure, allowing site reliability engineers (SRE) to focus on higher-value tasks.   Starting September 1, 2025, billing will begin for Azure SRE Agent. Please visit the Azure pricing page to see the pricing in your region. This article describes the billing model and example scenarios.  Overview of Azure SRE Agent Azure SRE Agent is a powerful tool that uses machine learning and advanced monitoring capabilities to proactively manage …  ( 33 min )

In certain scenarios, a storage account may contain a significant number of page blobs classified under the hot access tier that are infrequently accessed or retained solely for backup purposes. To optimise costs, it is desirable to transition these page blobs to the archive tier. However, as indicated in the following documentation - https://learn.microsoft.com/en-us/azure/storage/blobs/access-tiers-overview the ability to set the access tier is only available for block blobs; this functionality is not supported for append or page blobs. The Azure blob storage connector in Azure data factory is capable of copying blobs from block, append, or page blobs and copying data to only block blobs. https://learn.microsoft.com/en-us/azure/data-factory/connector-azure-blob-storage?tabs=data-factory#…  ( 24 min )

RFT Observability ("Auto-Evals") Reinforcement Finetuning (RFT) observability, now in public preview, provides real-time, in-depth visibility into your RFT job by automatically kicking off evaluation (“auto-evals”) that shows the detailed finetuning progress at each checkpoint.  What is RFT? RFT adjusts model weights using a reward model (grader) to score outputs against the reference data. The grader’s results are then used to reward or penalize the model response, steering the model’s reasoning direction and quality towards desired outcomes. The benefit of using RFT over supervised fine-tuning (SFT) is that RFT incorporates these reward signals real-time during training – but it’s slower and more expensive than SFT. Previously, customers could only evaluate RFT results after a training …  ( 28 min )

Fast Copy in Dataflows Gen2 is a game-changer to enhance the performance and cost-efficiency of your Dataflows Gen2. By leveraging the same optimized backend as the Copy Activity in data pipelines, Fast Copy significantly reduces data processing time and enhances cost efficiency. Fast Copy in Dataflows Gen2 (Generally Available) enabled by default in all newly created Dataflows … Continue reading “Boost Performance with Fast Copy in Dataflows Gen2 for Snowflake”  ( 6 min )

Simplified, workspace-governed ingestion without external dependencies COPY INTO and OPENROWSET from OneLake are now available in Preview for Microsoft Fabric Data Warehouse. With this release, you can load and query files directly from Lakehouse File folders, without relying on external staging storage, SAS tokens, or complex IAM configurations. This improvement reinforces Fabric’s vision of a … Continue reading “OneLake as a Source for COPY INTO and OPENROWSET (Preview)”  ( 6 min )

In April, we announced the preview of SQL Audit Logs for Microsoft Fabric Data Warehouse, giving organizations the power to track and retain critical warehouse events for enhanced visibility and control. Today, we’re taking that experience to the next level. We’re introducing an all-new, intuitive visual experience—making it easier than ever to configure, enable, and … Continue reading “Experience the New Visual SQL Audit Logs Configuration in Fabric Warehouse”  ( 6 min )

At the Build conference on May 21, 2024, we announced the general available of Personal Voice, a feature designed to empower customers to build applications where users can easily create and utilize their own AI voices (see the blog). Today we're thrilled to announce that Azure AI Speech Service has upgraded a new zero-shot TTS (text-to-speech) model, named “DragonV2.1Neural”. This new model delivers more natural-sounding and expressive voices, offering improved pronunciation accuracy and greater controllability compared to the earlier zero-shot TTS model. In this blog, we’ll present the new zero-shot TTS model audio quality, new features and benchmarks results. We’ll also share a guide for controlling pronunciation and accent using the Personal Voice API with the new zero-shot TTS model. …  ( 39 min )

This week in Model Mondays, we dive into the cutting-edge research happening at Azure AI Foundry Labs. From the MCP Server that makes it easy to experiment with new models and tools, to Magentic-UI that brings human-centered agent workflows to life, there’s a lot to unpack! Spotlight On: Research & Innovation in Azure AI Foundry Labs 1. What is this topic and why is it important? Research and innovation are the engines that drive progress in AI. This episode spotlights Azure AI Foundry Labs—a hub where Microsoft’s latest breakthroughs in models, agent frameworks, and developer tools are shared with the world. The MCP Server makes it easy for an yone to experiment with new models and tools, while Magentic-UI brings human-centered agent workflows to life. These projects help bridge the gap …  ( 37 min )

The Model Context Protocol (MCP) is transforming how AI systems interact with real-world applications. From intelligent assistants to real-time streaming, MCP is already being adopted by leading companies—and now is your chance to get ahead. Join us for a four-part technical series designed to give you practical, production-ready skills in MCP development, integration, and deployment. Whether you're a developer, AI engineer, or cloud architect, this series will equip you with the tools to build and scale MCP-based solutions. 📅 English edition - 6PM IST (India Standard Time) ✅ Register at MCP Bootcamp APAC Session TitleDate & Time (IST) Creating Your First MCP Server Learn the fundamental concepts of the protocol and test your implementation using official tools. August 28, 6:00 PM MC…  ( 23 min )

Here is the July 2025 release of the on-premises data gateway (version 3000.278).  ( 5 min )

We’re thrilled to announce the general availability (GA) of Autoscale Billing for Apache Spark in Microsoft Fabric — a serverless billing model designed to offer greater flexibility, transparency, and cost efficiency for running Spark workloads at scale. With this model now fully supported, Spark Jobs can run independently of your Fabric capacity and are billed … Continue reading “Autoscale Billing for Spark in Microsoft Fabric (Generally Available)”  ( 7 min )

Welcome to the July 2025 Fabric Feature Summary! This month’s update covers major events like the return of the Microsoft Fabric Community Conference in Vienna, and the 10th anniversary of Power BI. Key platform enhancements include new domain tags and updates to the default category in the OneLake catalog. You’ll also find highlights on data … Continue reading “Fabric July 2025 Feature Summary”  ( 12 min )

The Microsoft Fabric tooling ecosystem has evolved in recent months as both the Fabric CLI and the Terraform Provider for Microsoft Fabric have become generally available, providing new opportunities for automating Fabric Administration tasks for those already experienced with Terraform for declarative deployment of infrastructure as code. This short blog series will provide practical guidance … Continue reading “Terraform Provider for Microsoft Fabric: #3 Creating a workload identity with Fabric permissions”  ( 10 min )

Correlate signals, run advanced analytics, and perform forensic investigations from a single copy of data — without costly migrations or data silos. Detect persistent, low-and-slow attacks with greater visibility, automate responses using scheduled jobs, and generate predictive insights by combining Copilot, KQL, and machine learning.     Vandana Mahtani, Microsoft Sentinel Principal Product Manager shows how to uncover long-running threats, streamline investigations, and automate defenses — all within a unified, AI-powered SIEM experience. Store security data for up to 12 years. Perfect for long-term investigations and compliance. Check out our new data lake in Microsoft Sentinel. Streamline your data strategy. Send high-volume logs to the new low-cost data lake tier and control retent…  ( 40 min )

We’re happy to announce the preview of JSON Lines (JSONL) support in the OPENROWSET(BULK) function for Microsoft Fabric Data Warehouse and SQL endpoints for Lakehouses. The OPENROWSET(BULK) function allows you to query external data stored in the lake using well-known T-SQL syntax. With this update, you can now also query files in JSON Lines format, expanding the range of supported formats and simplifying access to semi-structured … Continue reading “JSON Lines Support in OPENROWSET for Fabric Data Warehouse and Lakehouse SQL Endpoints (Preview)”  ( 7 min )

As we continue to mature the Microsoft Fabric platform, we’re taking steps to streamline and simplify the experience for administrators and compliance teams who rely on audit logs. Beginning July 2025, we’re consolidating a number of redundant audit operations into a unified model that reflects our broader platform architecture. This change impacts how operations for … Continue reading “Standardizing Audit Operations for Warehouse, DataMarts and SQL Analytics Endpoint.”  ( 6 min )

We are continuously enhancing data agents in Fabric to deliver more powerful and flexible data experiences. In February of this year, we introduced a host of new improvements coming to the AI Skill—including support for additional Data Sources such as Eventhouse KQL and Semantic Models. Initially, integration to data sources was limited to sources with … Continue reading “Expanded Data Agent Support for Large Data Sources”  ( 6 min )

In the current digital environment, applications are expected to offer more than basic functionality, they must demonstrate intelligence, adaptability, and provide seamless user experiences. To remain competitive, organizations increasingly recognize that effective data utilization is fundamental to driving innovation. Data powers real-time insights and supports advanced agentic AI systems capable of reasoning, acting, and learning. … Continue reading “AI Ready Apps: From RAG to Chat – Interacting with SQL Database in Microsoft Fabric using GraphQL and MCP”  ( 17 min )

Getting Started Here is the original article written in Japanese. There are some great articles that explain how to integrate an MCP server built on Azure with a declarative agent created using Microsoft Copilot Studio. These approaches aim to extend the agent’s capabilities rather than defining a role. Here were some of the challenges we encountered: The agent's behavior can only be tested through the Copilot Studio. We want to use pro-code development tools. We can't choose LLM as the orchestrator—for example, there's no way to specify GPT-4o. The agent doesn’t always response same one which they would if we were prompting the LLM directly. These limitations got me thinking: why not build the entire agent myself? At the same time, I still wanted to take advantage of the familiar Micros…  ( 56 min )

Introduction Imagine a world where every second of your organization’s video content whether it’s a crucial training session, a product demo, or an expert-led seminar becomes instantly accessible, searchable, and actionable. What if, instead of sifting through hours of footage, you could surface key insights, create concise summaries, and deliver dynamic, avatar-driven presentations in minutes? In a landscape overwhelmed by video, it’s not about storing more, it’s about unlocking value from every frame.  Join me as we dive into the future of intelligent video analytics, where Microsoft Azure’s cutting-edge AI transforms raw footage into powerful knowledge assets. The Business Challenge Traditional video content management faces several critical limitations: Content Discovery: Finding spec…  ( 40 min )

Welcome to the July 2025 edition of the Fabric Influencers Spotlight, a recurring monthly post here to shine a bright light on the places on the internet where Microsoft MVPs & Fabric Super Users are doing some amazing work on all aspects of Microsoft Fabric. The Microsoft Fabric Community team has created the Fabric Influencers Spotlight to … Continue reading “Fabric Influencers Spotlight July 2025”  ( 8 min )

In today’s fast-paced software development cycles, manual testing often becomes a significant bottleneck. Our team was facing a growing backlog of test cases that required repetitive manual execution—running the entire test suite every sprint. This consumed valuable time that could be better spent on exploratory testing and higher-value tasks. We set out to solve this […] The post From Manual Testing to AI-Generated Automation: Our Azure DevOps MCP + Playwright Success Story appeared first on Azure DevOps Blog.  ( 25 min )

Introduction Modern systems must be modular and interoperable by design. Integration is no longer a feature, it’s a requirement. Developers are expected to build architectures that connect easily with third-party platforms, but too often, core systems are designed in isolation. This disconnect creates friction for downstream teams and slows delivery. At Microsoft, SaaS platforms like SAP SuccessFactors and Eightfold support Talent Acquisition by handling functions such as requisition tracking, application workflows, and interview coordination. These tools help reduce costs and free up engineering focus for high-priority areas like Azure and AI. The real challenge is integrating them with internal systems such as Demand Planning, Offer Management, and Employee Central. This blog post outlin…  ( 40 min )

Introduction Modern systems must be modular and interoperable by design. Integration is no longer a feature, it’s a requirement. Developers are expected to build architectures that connect easily with third-party platforms, but too often, core systems are designed in isolation. This disconnect creates friction for downstream teams and slows delivery. At Microsoft, SaaS platforms like SAP SuccessFactors and Eightfold support Talent Acquisition by handling functions such as requisition tracking, application workflows, and interview coordination. These tools help reduce costs and free up engineering focus for high-priority areas like Azure and AI. The real challenge is integrating them with internal systems such as Demand Planning, Offer Management, and Employee Central. This blog post outlin…  ( 40 min )

Introduction Modern systems must be modular and interoperable by design. Integration is no longer a feature, it’s a requirement. Developers are expected to build architectures that connect easily with third-party platforms, but too often, core systems are designed in isolation. This disconnect creates friction for downstream teams and slows delivery. At Microsoft, SaaS platforms like SAP SuccessFactors and Eightfold support Talent Acquisition by handling functions such as requisition tracking, application workflows, and interview coordination. These tools help reduce costs and free up engineering focus for high-priority areas like Azure and AI. The real challenge is integrating them with internal systems such as Demand Planning, Offer Management, and Employee Central. This blog post outlin…  ( 40 min )

Introduction Modern systems must be modular and interoperable by design. Integration is no longer a feature, it’s a requirement. Developers are expected to build architectures that connect easily with third-party platforms, but too often, core systems are designed in isolation. This disconnect creates friction for downstream teams and slows delivery. At Microsoft, SaaS platforms like SAP SuccessFactors and Eightfold support Talent Acquisition by handling functions such as requisition tracking, application workflows, and interview coordination. These tools help reduce costs and free up engineering focus for high-priority areas like Azure and AI. The real challenge is integrating them with internal systems such as Demand Planning, Offer Management, and Employee Central. This blog post outlin…  ( 40 min )

Azure Monitor logs are trusted by hundreds of thousands of organizations to monitor mission-critical workloads. But with such a diverse customer base, there’s no one-size-fits-all solution. That’s why we’re excited to announce a series of major advancements to Auxiliary Logs, the Azure Monitor plan that is designed for high volume logs, making it more cost-effective and improving the tools for use. Auxiliary Logs works in tandem with all other Azure Monitor tools including the more powerful Basic and Analytics Logs plans. Together, they are the one-stop-shop for all the logging needs of an organization. Auxiliary Logs is also one of the underlying services behind Microsoft Sentinel and the newly announced Sentinel data lake enabling customers to use the same data for both observability and…  ( 28 min )

Stay securely connected with rearchitected 5G design — including six smart-switching antennas, eSIM and Wi-Fi 7 — without relying on hotspots. As the first Surface Laptop to feature 5G, it enables enterprise-ready AI features for deeper insights, productivity boosts, and powerful local inferencing wherever work happens. Stay connected anywhere. The first Surface laptop with built-in 5G — supporting NanoSIM, eSIM, smart signal switching, and international roaming. See it here. High-performance AI experiences. Surface Laptop 5G is powered by Intel Core Ultra processors with AI Boost. Watch here. No IT setup required. Surface Laptop 5G can arrive business-ready with zero-touch deployment and managed 5G policies. Check it out. QUICK LINKS: 00:00 — Surface Laptop 5G for Business 00:28 — Buil…  ( 29 min )

How Contoso uses MQTT sensors, public weather feeds and Fabric Real-Time Intelligence to monitor smart buildings. Jointly authored by Alicia Li and Arindam Chatterjee Why Real-Time Stream Processing Matters In the age of AI, as organizations embrace intelligent systems and data-driven decision-making, the ability to act on data the moment it arrives is unlocking new … Continue reading “From Signals to Insights: Building a Real-Time Streaming Data Platform with Fabric Eventstream”  ( 8 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP Link part 1 Part 2 - What MCP is, including its architecture and core components Link part 2 Part 3 - A demo of MCP — including how to configure it so anyone can run it Part 3 - A demo of MCP using Visual Studio Code Link Part 3 Part 3B - A demo of MCP using GitHub Coding Agent  Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Link part 4 Part 5 - How to add OAuth authentication/authorization to our MCP Server Link part 5 Part 3B - MCP Demo with GitHub Coding Agent This MCP demo will use the GitHub Copilot Coding Agent. The Coding Agent is an autonomous, AI-powered software development agent integrated into GitHub. It can work ind…  ( 23 min )

Last month we introduced the SQL analytics endpoint Metadata Sync REST API in preview, and we’re excited to announce that this API is now Generally available (GA). With this API you can programmatically trigger a refresh of your SQL analytics endpoint to keep tables in sync with any changes made in your lakehouse, native and mirrored databases, … Continue reading “Refresh SQL analytics endpoint Metadata REST API (Generally Available)”  ( 6 min )

Celebrating Enhanced Flexibility in the WDK Sample We’re thrilled to announce a major step forward for the Workload Development Kit (WDK): developers can now transform our workload sample into their favorite programming languages. Whether you prefer Python, Java, Node.js, or Go, we’ve made it easier to start building innovative workloads with the tools and languages … Continue reading “Empowering Workload Developers with Language Choice and Simplicity”  ( 6 min )

Additional authors: Joanne Wong The preview of the integration between Fabric data agents and Microsoft Copilot Studio is now available, introducing a robust capability that enables agents to interoperate seamlessly across tools. This enhancement is designed to streamline the development, deployment, and scaling of intelligent agents across enterprise data environments, reinforcing automation and extensibility within … Continue reading “Fabric Data Agents + Microsoft Copilot Studio: A New Era of Multi-Agent Orchestration (Preview)”  ( 7 min )

The Microsoft Fabric tooling ecosystem has evolved in recent months as both the Fabric CLI and the Terraform Provider for Microsoft Fabric have become generally available, providing new opportunities for automating Fabric Administration tasks for those already experienced with Terraform for declarative deployment of infrastructure as code. This short blog series will provide practical guidance … Continue reading “Terraform Provider for Microsoft Fabric: #2 Using MCP servers and Fabric CLI to help define your fabric resources”  ( 9 min )

It's nice to meet you virtually! We are a team consisting of Tamara Gaidar, a data scientist, and Fady Copty, a researcher. We both work at Microsoft Security. In the upcoming blog posts, we will share insights our team has gained from building AI agents that can be leveraged for any agentic-ai application. We hope you enjoy reading and that it adds value to your day-to-day job. Cheers! TL;DR In this blog we provide a practical guide for transitioning agentic AI projects from research to production. We will be writing it in several parts. In part one we will explore how to effectively design agentic AI systems - AI agents that can reason over and generate unstructured data - by emphasizing the importance of clearly defining the problem and understanding why AI is needed. We will outline a …  ( 29 min )

Integrating Azure Function Apps into your Azure Data Factory (ADF) workflows is a common practice. To enhance security beyond the use of function API keys, leveraging managed identity authentication is strongly recommended. Given the fact that many existing guides were outdated with recent updates to Azure services, this article provides a comprehensive, up-to-date walkthrough on configuring managed identity in ADF to securely call Function Apps. The provided methods can also be adapted to other Azure services that need to call Function Apps with managed identity authentication. The high level process is: Enable Managed Identity on Data Factory Configure Microsoft Entra Sign-in on Azure Function App Configure Linked Service in Data Factory Assign Permissions to the Data Factory in Azure …  ( 23 min )

A Java thread dump is a snapshot of the current state of all the threads that are part of a Java process. It provides a lot of valuable information about the Java process like its current execution point, stack trace, thread priority, and the status of the thread (running, waiting, or blocked).Thread dumps are especially useful for diagnosing and troubleshooting performance issues, deadlocks, and unresponsive applications in a Java application. They are crucial for understanding the behavior of an application at a certain point in time and can provide insights into the application's performance that would be difficult to obtain in any other way.   In Azure App Service, we provide ability to capture Java thread dump of your Java application by using "Diagnostic & Solve Problem". The tool is…  ( 25 min )

A search job is an asynchronous query that runs on any data in your Log Analytics workspace, including data from the long-term retention, making the results available for further queries in a new Analytics table within your workspace.  To efficiently search massive datasets, Search Job divides queries into smaller time-based segments, processes them in parallel, and returns the results. This approach optimizes scalability and enables reliable analysis, even over petabytes of data.  We’re excited to announce significant enhancements to Search Jobs, designed to make large-scale data exploration faster, easier, and more efficient.  What’s New in Search Job  Our latest update includes several powerful improvements:  Intuitive and streamlined UI experience for faster and simpler setup.  Cost e…  ( 24 min )

We are pleased to announce that Azure VMware Solution is now available in Spain Central. Now in 35 Azure regions, Azure VMware Solution empowers you to seamlessly extend or migrate existing VMware workloads to Azure without the cost, effort or risk of re-architecting applications or retooling operations.  Azure VMware Solution supports: Rapid cloud migration of VMware-based workloads to Azure without refactoring. Datacenter exit while maintaining operational consistency for the VMware environment. Business continuity and disaster recovery for on-premises VMware environments. Attach Azure services and innovate applications at your own pace. Includes the VMware technology stack and lets you leverage existing Microsoft licenses for Windows Server and SQL Server. For updates on current and upcoming region availability, visit the product by region page here. Streamline migration with new offers and licensing benefits, including a 20% discount. We recently announced the VMware Rapid Migration Plan, where Microsoft provides a comprehensive set of licensing benefits and programs to give you price protection and savings as you migrate to Azure VMware Solution. Azure VMware Solution is a great first step to the cloud for VMware customers, and this plan can help you get there. Learn More  ( 19 min )

Read/Write to SQL database in Fabric using Python Notebook You created your SQL database in Fabric because you wanted to take full advantage of all the existing workloads without worrying about integration pain points. Python notebook integration with SQL was highly requested. We are thrilled to share you can seamlessly use a python notebook to … Continue reading “Connect to your SQL database in Fabric using Python Notebook”  ( 6 min )

Building on the excitement of our public preview, we’re excited to announce the latest updates to the Migration Assistant for Fabric Data Warehouse, making it even easier, faster, and more reliable to move your data to Fabric. The Migration Assistant simplifies what can be a complicated process by automatically taking care of it. Since its … Continue reading “Fabric Data Warehouse Migration Assistant: Better, Faster, More Reliable”  ( 6 min )

Copy Job is now the preferred solution in Microsoft Fabric Data Factory for simplified data movement, whether you’re moving data across clouds, from on-premises systems, or between different services. With robust support for both batch and incremental copying, it offers the flexibility to tackle a wide range of scenarios with ease. We continuously improve Copy … Continue reading “Simplifying Data Ingestion with Copy job – Copy Data from Database View,  Sample dataset and New Connectors”  ( 6 min )

As generative AI adoption accelerates, more professionals are recognizing the limitations of basic Retrieval-Augmented Generation (RAG). Many point out that traditional RAG provides only superficial analysis and that search results often appear as isolated points, rather than forming holistic insights. Most RAG implementations rely on single-query searches and summarization, which makes it difficult to explore information in depth or to perform repeated validation. These limitations become especially clear in complex enterprise scenarios. Deep Research is designed to address these challenges. Unlike RAG, Deep Research refers to advanced capabilities where AI receives a user’s query and collects and analyzes information from a variety of perspectives to generate detailed reports. This appro…  ( 46 min )

Hello Java customers! We are happy to announce the latest July 2025 patch & security update release for the Microsoft Build of OpenJDK. Download and install the binaries today. OpenJDK 21.0.8 OpenJDK 17.0.16 OpenJDK 11.0.28 Check our release notes page for details on fixes and enhancements. The source code of our builds is available now […] The post Java OpenJDK July 2025 Patch & Security Update appeared first on Microsoft for Java Developers.  ( 22 min )

Overview  As part of an upcoming industry-wide change, DigiCert, the Certificate Authority (CA) for Azure App Service Managed Certificates (ASMC), is required to migrate to a new validation platform to meet multi-perspective issuance corroboration (MPIC) requirements.   While most certificates will not be impacted by this change, certain site configurations and setups may prevent certificate issuance or renewal starting July 28, 2025.   What Will the Change Look Like?  For most customers: No disruption. Certificate issuance and renewals will continue as expected for eligible site configurations.  For impacted scenarios: Certificate requests will fail (no certificate issued) starting July 28, 2025, if your site configuration is not supported. Existing certificates will remain valid until th…  ( 35 min )

As AI agents become central to how we build and interact with modern productivity apps, developers need tools that are flexible, standard-based, and deeply integrated with AI. That’s where the Microsoft 365 Agents Toolkit MCP Server comes in.  This is a local Model Context Protocol (MCP) server that acts as a bridge between your AI coding agents and […] The post Build Smarter with the Microsoft 365 Agents Toolkit MCP Server appeared first on Microsoft 365 Developer Blog.  ( 24 min )

This post walks through how to implement a “build once, deploy everywhere” pattern using Azure Developer CLI (azd) that provisions environment-specific infrastructure and promotes applications from dev to prod with the same build artifacts. You’ll learn how to use conditional Bicep deployment, environment variable injection, package preservation across environments, and automated CI/CD promotion from development […] The post Azure Developer CLI: From Dev to Prod with One Click appeared first on Azure DevOps Blog.  ( 24 min )

Overview Microsoft Fabric is officially sunsetting Default Semantic Models. This change is part of our ongoing efforts to simplify and improve the manageability, deployment, and governance of Fabric items such as warehouse, lakehouse, SQL database, and mirrored databases. Why the Change? Default Semantic Models were initially designed to provide a lightweight, out-of-the-box experience—automatically generating models … Continue reading “Sunsetting Default Semantic Models – Microsoft Fabric”  ( 7 min )

Fabric offers a wide variety of data-science capabilities, from automated machine learning with FLAML to batch inferencing with the SynapseML PREDICT function. We’re pleased to announce that ML models can now serve real-time predictions from secure, scalable, and easy-to-use online endpoints. In addition to generating batch predictions in Spark, you can use endpoints to bring … Continue reading “Serve real-time predictions seamlessly with ML model endpoints”  ( 6 min )

We’re thrilled to share a series of exciting updates and upcoming enhancements to the SQL analytics endpoint in Microsoft Fabric. These improvements are designed to make your experience more powerful and reliable. From Preview to GA – Metadata sync REST API Last month, we introduced the SQL analytics endpoint metadata sync REST API in preview, … Continue reading “What’s new and coming soon in SQL analytics endpoint in Fabric”  ( 6 min )

Microsoft Fabric offers native Git integration and deployment pipelines to facilitate version control, collaboration, and automated releases for workspace items like user data functions. This guide explains how to set up and manage Git integration for user data functions within a Fabric workspace. • Workspace preparation and Git linking: Users start by selecting or creating a Fabric workspace containing user data functions, then enable Git integration via workspace settings by connecting to a Git provider and repository branch, optionally specifying a folder for organization. • Branching strategy configuration: Teams are advised to adopt branching strategies such as main/develop, feature, and release branches, along with pull request and code review policies to maintain code quality and collaboration. • Managing user data functions in Git: Each data function is stored in a function_app.py file; users clone the repository locally, edit or add functions, and update the definition.json file to reflect new functions and required libraries like numpy. • Committing, syncing, and publishing changes: After committing changes in VS Code, users sync with the Fabric portal, update the function via source control, and publish to deploy the new or updated functions, making them available for invocation.  ( 8 min )

When you deploy AKS, you deploy the control plan, which is managed by Microsoft, and one or more node pools, which contain the worker nodes used to run your Kubernetes workloads. These node pools are usually deployed as Virtual Machine Scale Sets. These scale sets are visible in your subscription, but generally you would not make changes to these directly, as they will be managed by AKS and all of the configuration and management of these is done through AKS. However, there are some scenarios where you do need to make changes to the underlying node configuration to be able to handle the workloads you need to run. Whilst you can make some changes to these nodes, you need to make sure you do it in a supported manner, which will be applied consistently to all your nodes. An example of this re…  ( 38 min )

In the rapidly evolving landscape of AI application development, the ability to orchestrate multiple intelligent agents has become crucial for building sophisticated, enterprise-grade solutions. While individual AI agents excel at specific tasks, complex business scenarios often require coordination between specialized agents running on different platforms, frameworks, or even across organizational boundaries. This is where the […] The post Guest Blog: Building Multi-Agent Solutions with Semantic Kernel and A2A Protocol appeared first on Semantic Kernel.  ( 29 min )

Project Overview The system replicates a developer's workflow by autonomously launching VS Code, configuring the environment, and generating code via GitHub Copilot Agent mode. This automation is ideal for scenarios where: Codebase is not hosted on GitHub – GitHub Agents on github.com are not applicable. GitHub Codespaces cannot access the code – VS Code in the browser is ruled out. Desktop automation is required – Must work with VS Code desktop client on a local computer. Key Innovation: CUA Model + PyAutoGUI Integration The project demonstrates the synergy between CUA model and PyAutoGUI: PyAutoGUI: Executes desktop actions (launching apps, typing, clicking). CUA Model: Analyzes screenshots and determines next actions. Deterministic Outcomes: Enables autonomous detection and correctio…  ( 22 min )

Solution ideas The original framework introduced several Logic App and Function App patterns for SQL BPA, Update Manager, Cost Management, Anomaly Detection, and Smart Doc creation. In this article we add two Copilot Studio Agents, packaged in the GitHub repository Microsoft Azure AI for Operation Framework, designed to be deployed in a dedicated subscription (e.g., OpenAI-CoreIntegration): Copilot FinOps Agent – interactive cost & usage analysis Copilot Update Manager Agent – interactive patch status & one-time updates Architecture   Copilot FinOps Agent A Copilot Studio agent that lets stakeholders chat in natural language to retrieve, compare, and summarise cost data—without leaving Teams. Dataflow #StageDescription Initial TriggerUser message (Teams / Copilot Studio web) invoke t…  ( 27 min )

With the recent release of our new Availability Zone features, a common question we've been getting from our customers is: "How do I know if my existing App Service supports Availability Zones?" While this information is documented in the App Service Resiliency Microsoft Learn Docs, in this blog post I will try and walk you through some essential aspects of enabling Availability Zones in Azure App Service. This article outlines how to confirm if your App Service plan supports Availability Zones, guides you through upgrading from Free, Basic, or Standard tiers to Premium SKUs, and describes the process for enabling this feature, including creating new, properly configured plans. The overview also addresses specific considerations for App Service Environments, along with common issues and so…  ( 44 min )

Overview Azure VMware Solution is a VMware validated first party Azure service from Microsoft that provides private clouds containing VMware vSphere clusters built from dedicated bare-metal Azure infrastructure. It enables customers to leverage their existing investments in VMware skills and tools, allowing them to focus on developing and running their VMware-based workloads on Azure. At Microsoft, we’re continuously evolving our services based on customer feedback and Azure VMware Solution is no exception. As a fully Microsoft-managed service, Azure VMware Solution takes care of the end-to-end lifecycle management of your VMware environment, from ESXi host patching to vCenter Server and NSX upgrades. This ensures your private cloud stays secure, compliant, and up-to-date.  Historically, p…  ( 25 min )

Shortcut‑based AI transformations in Microsoft Fabric convert raw text files into governed Delta Lake tables within minutes, removing the need for complex data‑integration pipelines and significantly reducing time to insight. Why adopt AI transformations? Supported AI transforms Transform Purpose Summarization Generates concise summaries from long-form text. Translation Translates text between supported languages. Sentiment analysis Labels text … Continue reading “Accelerating Insights from Unstructured Text with AI Powered OneLake Shortcut Transformations”  ( 6 min )

Today, we are thrilled to announce the General Availability of the Geo-replication feature for Azure Event Hubs, now available in both Premium and Dedicated tiers. This milestone marks a significant enhancement in our service, providing our customers with robust business continuity and disaster recovery capabilities – ensuring high availability for their mission-critical applications.  The Geo-replication feature allows you to replicate your Event Hubs data across multiple regions either synchronously or asynchronously, ensuring that your data remains accessible in the event of maintenance activities, regional degradation, or a regional outage. With Geo-replication, you can seamlessly promote a secondary region to a primary, minimizing downtime and ensuring business continuity.        …  ( 21 min )

What Will We Cover? In this interactive AMA, you'll learn how to: Expose Azure API Management instances as MCP servers, enabling remote access to your APIs using AI combined with Model Context Protocol. Configure API Management policies to enhance your MCP servers with enterprise-grade capabilities such as rate limiting, authentication, and centralized monitoring. Why This Matters Model Context Protocol (MCP) bridges the gap between AI agents and the real-world data they need to be effective. By integrating MCP with Azure API Management, developers can expose tools to their AI agents while enforcing consistent policies and security standards. Whether you’re deploying custom tools or remote services, this AMA will show you how Azure API Management can be your go-to platform for controlling and scaling MCP access. How to Join Register to Join the Azure AI Foundry Discord Community EventSee the events channel 📅 Monday, July 21st, 2025⏰ 10:00 AM Pacific Time (UTC−07:00) Event Highlights Learn how to expose MCP servers through Azure API Management See how to configure and test policies such as rate limiting Get answers directly from Microsoft product managers and engineers Connect with fellow developers building with MCP, Azure API Management, and Azure AI Foundry Get a Head Start Before the event, check out the documentation to learn how to Expose a REST API in API Management as an MCP server, view the Build and protect MCPs faster with governance in Azure API Manager session from Build 2025, or explore the AI-Gateway labs on GitHub and learn how to use APIM and MCP in the MCP for Beginners course. Don’t miss this opportunity to deepen your understanding of API Management and MCP integration—and get your questions answered live!  ( 21 min )

We are pleased to announce that Eventstream’s Confluent Cloud for Apache Kafka streaming connector now supports decoding data from Confluent Cloud for Apache Kafka topics that are associated with a data contract in Confluent Schema Registry. The Challenge with Schema Registry Encoded Data The Confluent Schema Registry serves as a centralized service for managing and … Continue reading “Decoding Data with Confluent Schema Registry Support in Eventstream (Preview)”  ( 7 min )

Shortcut‑based AI transformations in Microsoft Fabric convert raw text files into governed Delta Lake tables within minutes, removing the need for complex data‑integration pipelines and significantly reducing time to insight. Why adopt AI transformations? How to set up an AI-powered shortcut transformations 3. Monitor progressStatus can be observed under Manage shortcut Practical implementation patterns All these … Continue reading “Accelerating Insights from Unstructured Text with AI‑Powered OneLake Shortcut Transformations in Microsoft Fabric (Public Preview)”  ( 3 min )

Introducing multiple-schema inferencing in Eventstream! This feature empowers you to work seamlessly with data sources that emit varying schemas by inferring and managing multiple schemas simultaneously. It eliminates the limitations of single-schema inferencing by enabling more accurate and flexible transformations, preventing field mismatches when switching between Live and Edit modes, and allowing you to view … Continue reading “Enhancing Data Transformation Flexibility with Multiple-Schema Inferencing in Eventstream (Preview)”  ( 8 min )

Microsoft Fabric has introduced new features for its User Data Functions (UDFs), enhancing Python-based data processing capabilities within the platform. These updates include support for asynchronous functions and the use of pandas DataFrame and Series types for input and output, enabling more efficient handling of large-scale data. • Async function support: Developers can now write async functions in Fabric UDFs to improve responsiveness and efficiency, especially for managing high volumes of I/O-bound operations, such as reading files asynchronously from a Lakehouse. • Pandas DataFrame and Series integration: UDFs can accept and return pandas DataFrames and Series, allowing batch processing of rows with improved speed and performance in data analysis tasks. An example function calculates total revenue by driver using pandas groupby operations. • Usage in notebooks: These functions can be invoked directly from notebooks using pandas objects, facilitating efficient aggregation and analysis of large datasets interactively within Microsoft Fabric. • Getting started and benefits: Users can enable these features by updating the fabric-user-data-functions library to version 1.0.0. The enhancements reduce I/O operations, enable concurrent task handling, and improve performance on datasets with millions of rows.  ( 7 min )

This post was generated with AI help and human revision & review. To learn more about our motivation and workflows, please refer to this document in our Model Mondays website. About Model Mondays Model Mondays is a weekly series designed to help you build your Azure AI Foundry Model IQ, one week at a time. Here’s what to expect: 5-Minute Highlights – Quick updates on Azure AI models and tools (Mondays) 15-Minute Spotlight – A deeper look at a key model, protocol, or feature (Mondays) 30-Minute AMA – Friday Q&A with experts from Monday’s episode Whether you’re just starting out or already working with models, this series is your chance to grow with the community. Quick links to explore: Register for Model Mondays Watch Past Episodes Join the AMA on July 18 Visit the Discussion Forum Sp…  ( 26 min )

If you want to know about Azure, the work we are doing in partnership with VMware by Broadcom, or have a conversation about your VMware workloads, stop by our booth #107! This year Microsoft will have several break-out sessions on a variety of topics where you will discover how Microsoft can streamline your VMware workloads’ migration to the cloud. Learn how to maximize your existing on-premises investments and fast path your applications into an AI-innovation platform. We’ll reveal the quickest, least disruptive migration path, and share enticing offers that make this an easy decision. Learn how transitioning from on-premises to Azure will give your business a competitive edge. As you are building out your VMware Explore schedule don't miss these Micrsoft sessions: Monday, August 25th9:00-9:45 AM From On-Prem to Intelligent Cloud with Azure VMware Solution    CLOB1948LVS  45-min. Breakout Session Monday, August 25th 12:45-1:30 PM Building End-to-End Networking with Azure VMware Solution   CLOB1950LVS  45-min. Breakout Session Monday, August 25th 2:00-3:30 PM Azure & AI: How to unleash the power of Azure VMware Solution   CLOT1951LVS  90-min. Tutorial Tuesday, August 26th 10:30-11:15 AM Enterprise Security and Migration with Azure VMware Solution   CLOB1952LVS  45-min. Breakout Session Tuesday, August 26th1:00-1:45 PM Build Resilient VCF Private Clouds: Azure VMware Solution (AVS) + BCDR & Storage   CLOB1953LVS  45-min. Breakout  Wednesday, August 27th 11:00-11:30 AM Ask Me Anything About Azure VMware Solution   CLOM1949LVS  Meet the Expert Roundtable Wednesday, August 27th3:30-4:00 PM Ask Me Anything About Azure VMware Solution   CLOM1947LVS  Meet the Expert Roundtable  ( 20 min )

Custom Engine Agents now generally available—build and integrate your own AI into the flow of work Microsoft 365 Copilot is redefining how people interact with AI—embedding it directly into the flow of work as the intuitive, natural interface for agents: the ‘UI for AI’ As Copilot becomes the interface for AI in the workplace, we’re […] The post Bring your own agents into Microsoft 365 Copilot appeared first on Microsoft 365 Developer Blog.  ( 26 min )

In many use cases, using remote MCP servers is not uncommon. In particular, if you're using Azure API Management (APIM), Azure API Center (APIC) or Copilot Studio in Power Platform, integrating with remote MCP servers is inevitable.   MCP (Model Context Protocol) is a protocol used for defining server endpoints that support AI agent workflows, often in Copilot scenarios.   Background   A remote MCP server is basically an HTTP API app that exposes an endpoint of either /mcp or /sse depending on the spec the servers support. This endpoint should be accessible from Copilot Studio through a custom connector, APIM or APIC. The easiest and simplest way of exposing this endpoint is to provide an OpenAPI doc. The OpenAPI doc is very simple because it has only one endpoint, and the structure is a…  ( 28 min )

Purpose of the document provide overview of AI Foundry and how it can be set up and organised for at scale for an enterprise.  This document should be considered as guidance and recommendations, but individual organisations should treat and consider other factors such as security, policy, governance and number of business units. AI Foundry Resource: Azure AI Foundry is Microsoft’s unified platform for building, customising, and managing AI applications and agents—designed to accelerate innovation and operationalise AI at scale. It brings together: Data, models, and operations into a single, integrated environment. A developer-first experience with native support for GitHub, Visual Studio, and Copilot Studio. A low-code portal and code-first SDKs/APIs for flexibility across skill levels. …  ( 43 min )

Azure Custom Vision is an easy-to-use yet powerful cognitive service from Microsoft that empowers developers to build, train, and deploy custom image classification or object detection models - all without being machine learning experts. Through a simple web interface or SDKs, you can upload images, tag them with meaningful labels, and train a model that understands your specific use case with the help of transfer learning. Once trained, you might realize you need access to those tagged images again - maybe for auditing, retraining on different platforms, migrating data, or simply backing them up. Unfortunately, the Custom Vision portal doesn’t offer a “Download All Tagged Images” button. Thankfully, Microsoft provides a RESTful Get Tagged Images API, a hidden gem that enables you to reco…  ( 39 min )

The artificial intelligence landscape presents unprecedented opportunities for Java developers. With decades of experience building robust enterprise applications using Spring, Quarkus and Jakarta EE, Java developers are uniquely positioned to create sophisticated AI solutions. The emergence of specialized AI libraries like LangChain4j enables them to leverage their existing expertise while building secure, scalable, and maintainable […] The post Microsoft and LangChain4j: A Partnership for Secure, Enterprise-Grade Java AI Applications appeared first on Microsoft for Java Developers.  ( 23 min )

Instead of switching between logs, PowerShell, and spreadsheets, Security Copilot centralizes insights for faster, more focused action.  Resolve compromised accounts, uncover ownerless or high-risk apps, and tighten policy coverage with clear insights, actionable recommendations, and auto-generated policies. Strengthen security posture and reclaim time with a smarter, more efficient approach powered by Security Copilot.       Diana Vicezar, Microsoft Entra Product Manager, shares how to streamline investigations and policy management using AI-driven insights and automation. Skip the scripting.  Ask questions in plain language and get back policy and risk insights in seconds. Microsoft Entra now has built-in AI with Security Copilot. Stay ahead of threats.  Use AI to track auth changes, …  ( 38 min )

We’re excited to announce the General Availability of Managed Private Endpoints (MPE) in Fabric Eventstream. This network security feature allows you to stream data from Azure resources to Fabric over a private and secure network without the complexity of manual network configurations. Why Network Security Matters for Streaming As organizations increasingly adopt real-time data streaming … Continue reading “Secure Data Streaming with Private Endpoints in Eventstream (Generally Available)”  ( 6 min )

Azure and Fabric Events offer a powerful capability within Real-Time Intelligence that enables you to ingest system events that are generated in Microsoft Fabric and Azure to deliver them to consumers in Microsoft Fabric like Activator for setting event-based triggers or Eventstream to stream and process events to other destinations. Permission model To subscribe to Azure and Fabric events … Continue reading “Permission model improvements for Azure and Fabric Events”  ( 7 min )

The Microsoft Fabric tooling ecosystem has evolved in recent months as both the Fabric CLI and the Terraform Provider for Microsoft Fabric have become generally available, providing new opportunities for automating Fabric Administration tasks for those already experienced with Terraform for declarative deployment of infrastructure as code. This short blog series will provide practical guidance … Continue reading “Terraform Provider for Microsoft Fabric: #1 Accelerating first steps using the CLIs”  ( 10 min )

Project Overview We are a team of Imperial College Students committed to improving campus life through innovative multi-agent solutions. CampusSphere leverages Microsoft Azure AI capabilities to automate core university campus services. We created an end-to-end solution that allows both students and staff to access a multi-agent framework for room/gym booking, attendance tracking, calendar management, IoT monitoring and more. 🔭 Our Initial Vision: Reimagining Campus Technology When our team at Imperial College London embarked on the CampusSphere project as part of Microsoft's Agentic Campus initiative, we had one clear ambition: to create an intelligent campus ecosystem that would fundamentally change how students, faculty, and staff interact with university services. The inspiration came…  ( 31 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP Link part 1 Part 2 - What MCP is, including its architecture and core components Link part 2 Part 3 - A demo of MCP — including how to configure it so anyone can run it Link part 3 Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Link part 4 Part 5 - How to add OAuth authentication/authorization to our MCP Server  Part 5 ... MCP Authentication / Authorization In the previous section we covered the development of an MCP Server. In this section we shall extend the server code so that it supports authentication / authorization. Objectives Authentication and authorization are essential for both the MCP service provider and the user. Each…  ( 40 min )

Broadcom has released a new Critical Security Advisory, VMSA-2025-0013 with a CVSS base score range of 7.1 to 9.3. With Microsoft’s commitment to the security of our platform and our improved lifecycle management process, we were able to quickly assemble a global team to work on the acceleration and validation of the ESXi 8.0 U3f + Hot Patch (VAIO bug fix) Build 24797835 security patch . We have nearly finished qualifying the security patch that will mitigate VMSA-2025-0013 across our fleet. As a result, with the public release of this vulnerability we expect to be able to patch your existing Azure VMware Solution infrastructure next week. We are committing to completing the remediation within 30-days. Microsoft will communicate the scheduled date of patching over the next three weeks. Any…  ( 22 min )

In August 2025, Microsoft Fabric will introduce workspace access limits to improve service quality, reliability, and to encourage workspace access control hygiene. This limit will be permanent once it is rolled out – each Fabric & Power BI workspace will be limited to a maximum of 1,000 users or groups in workspaces roles (Admin, Member, … Continue reading “Introduction of access limits in a Fabric workspace”  ( 6 min )

We are thrilled to announce the general availability of Mirroring for Azure Databricks Unity Catalog in Microsoft Fabric—a secure, high-performance integration that provides seamless access to Azure Databricks tables from Fabric. With Fabric and Azure Databricks, we are building the future of data platforms on a lakehouse foundation, powered by open data formats, full interoperability, … Continue reading “Unified by design: mirroring Azure Databricks Unity Catalog to Microsoft OneLake in Fabric (Generally Available)”  ( 9 min )

The world of AI agents is evolving rapidly, with new protocols and frameworks emerging to enable sophisticated multi-agent communication. Google's Agent-to-Agent (A2A) protocol represents one of the most promising approaches for building distributed AI systems that can coordinate tasks across different platforms and services. I'm excited to share how you can leverage Azure App Service to build, deploy, and scale A2A applications. Today, I'll walk you through a practical example that combines Microsoft Semantic Kernel with the A2A protocol to create an intelligent travel planning assistant. What We Built: An A2A Travel Agent on App Service I've taken an existing A2A travel planning sample and enhanced it to run seamlessly on Azure App Service. This demonstrates how A2A concepts can be adapt…  ( 32 min )

We're announcing a new tutorial that helps developers migrate solutions from EWS to Microsoft Graph using AI and GitHub Copilot. The post Supercharge your EWS migration with AI and GitHub Copilot appeared first on Microsoft 365 Developer Blog.  ( 23 min )

You can now interactively debug Hyperlight guest micro-VMs. Attach the GNU Debugger at runtime to step through the code. The post Hyperlight: Debugging hardware-protected guests appeared first on Microsoft Open Source Blog.  ( 16 min )

In a world where foundation models are increasingly accessible, the real magic happens when developers make them their own. Fine-tuning is no longer a niche capability, it’s becoming a core skill for developers who want to build AI that’s faster, smarter, and more aligned with their users scaling expert knowledge across their organizations. Over the past few months, we’ve seen something remarkable: a growing community of builders, tinkerers, and innovators coming together to push the boundaries of what fine-tuning can do making a powerful difference for everyday organizations.  A Community Making a Big Impact: Customer Stories At Build 2025, we saw firsthand how much the landscape has shifted. Just a year ago, many teams were still relying solely on prompt engineering or retrieval-augmente…  ( 29 min )

As organizations adopt Azure VMware Solution (AVS) to modernize their infrastructure, the need for repeatable, script-driven deployment patterns becomes increasingly important. Whether you're preparing for a new AVS deployment or enhancing an existing AVS environment, automation can significantly reduce manual effort, improve consistency, and accelerate readiness for workload migration.   AVS Landing Zone already provides a rich repository of Terraform, Azure Resource manager (ARM), Bicep and PowerShell scripts to automate deployment as well as management operations. For example, we recently added a collection of PowerShell-based automation scripts that simplify the deployment of key AVS components. This article highlights three such scripts that help accelerate AVS migrations: Jumpbox De…  ( 30 min )

Microsoft’s AI-powered code review assistant has transformed pull request workflows by automating routine checks, suggesting improvements, and enabling conversational Q&A, leading to faster PR completion, improved code quality, and enhanced developer onboarding. Its seamless integration and customizability have driven widespread adoption within Microsoft The post Enhancing Code Quality at Scale with AI-Powered Code Reviews appeared first on Engineering@Microsoft.  ( 27 min )

After the successful release of Public Preview of mssql-python driver, we’re thrilled to announce a major milestone for the mssql-python driver: full support for all three major operating systems—Windows, macOS, and Linux. This release marks a significant leap forward in our mission to provide seamless, performant, and Pythonic connectivity to Microsoft SQL Server and the […] The post Announcing Full Cross-Platform Support for the mssql-python Driver appeared first on Microsoft for Python Developers Blog.  ( 24 min )

Coauthor: Ramachandran G, RTI Engineering Manager, Microsoft In today’s data-driven world, organizations rely heavily on real-time insights to monitor systems, detect anomalies, and make informed decisions. One of the key challenges in achieving this is efficiently ingesting and transforming log data from diverse sources into a format that can be analysed instantly. Real-Time Intelligence Real-time … Continue reading “Ingest Logs using Logstash into Real-Time Intelligence”  ( 7 min )

The Preview of Cosmos DB in Microsoft Fabric, is now available to all users. Following its initial announcement at Microsoft Build 2025, several new capabilities have been added to improve data workflows. With this release, you can seamlessly access and analyze your operational data across the Fabric ecosystem. Leverage Real-Time Intelligence, Copilot-powered Power BI, and … Continue reading “Announcing Cosmos DB in Microsoft Fabric Featuring New Capabilities! (Preview)”  ( 8 min )

Microsoft Fabric now offers a preview of support for Microsoft Entra service principals when using Amazon S3 Shortcuts. This feature allows the use of Entra service principals to securely access S3 buckets without the need for long-term AWS access keys. Previously, S3 shortcuts required access keys. With this update, organizations can authorize access using Microsoft … Continue reading “Access Amazon S3 Shortcuts Securely and Seamlessly with Microsoft Entra Service Principals (Preview)”  ( 6 min )

Overview Model Context Protocol (MCP) is an open, standardized protocol that allows Large Language Models (LLMs) to interact with external tools and data sources in a consistent, extensible way. It gives models access to capabilities beyond their training data—such as calling APIs or querying databases. MCP builds on the idea of function calling, which lets LLMs invoke external operations by generating JSON payloads that match predefined schemas. While function calling was initially tied to proprietary formats (like OpenAI’s schemas), MCP unifies these patterns under a common JSON-RPC-based protocol. This makes it easier for developers to write tools once and expose them to many LLMs, regardless of vendor. In short: Function calling gave LLMs actions. MCP gives those actions structure, di…  ( 41 min )

As AI agents become more sophisticated, the need for seamless integration with powerful cloud-based tools grows. With the Azure AI Foundry SDK and MCP (Model Context Protocol) tools, creates a dynamic duo that empowers developers to build, deploy, and manage intelligent agents with ease. Solution Overview The AI-Foundry-Agent-MCP GitHub repository provides a hands-on solution for integrating MCP tools with the Azure AI Foundry SDK. This setup allows developers to: Access and deploy state-of-the-art models from Azure AI Foundry. Use MCP tools to manage model context, knowledge bases, and evaluation pipelines. Rapidly prototype and scale AI agent solutions in a cloud-native environment. Getting Started The repo includes a step-by-step guide to get your environment up and running: Clone th…  ( 23 min )

The July 2025 release includes TBA and more! The post Python in Visual Studio Code – July 2025 Release appeared first on Microsoft for Python Developers Blog.  ( 23 min )

In this blog post, we will describe how to train an ML model in Fabric Spark notebook, save it in Fabric’s models registry and use it for scoring new streaming data by Fabric Eventhouse via update policy in real time. We describe a typical workflow that can be implemented to monitor cloud resources or IoT … Continue reading “ML Model Scoring in Fabric Eventhouse via Update Policy”  ( 11 min )

Effortlessly read Delta Lake tables using Apache Iceberg readers Microsoft Fabric is a unified, SaaS data and analytics platform designed for the era of AI. All workloads in Microsoft Fabric use Delta Lake as the standard, open-source table format. With Microsoft OneLake, Fabric’s unified SaaS data lake, customers can unify their data estate across multiple … Continue reading “New in OneLake: Access your Delta Lake tables as Iceberg automatically (Preview)”  ( 6 min )

A new feature has been added to Eventstream—the SQL Operator—which enables real-time data transformation within the platform. Whether you’re filtering, aggregating, or joining data streams, or handling complex data transformation needs like conditional logic, nested expression, string manipulation etc. SQL Operator gives you the flexibility and control to craft custom transformations using the language you … Continue reading “From Clicks to Code: SQL Operator under Fabric Eventstream (Preview)”  ( 7 min )

For enterprise customers, SAP represents one of the most valuable data sources. Integration of SAP data with other non-SAP data estates is a frequent requirement of Microsoft Fabric customers. Consequently, we are continually working to expand the options for SAP data integration to ensure that every customer has a viable solution tailored to their specific … Continue reading “What’s new with SAP connectivity in Microsoft Fabric – July 2025”  ( 9 min )

Have you ever wondered how to give AI assistants access to your organization’s data in a clean, structured way? The Model Context Protocol (MCP) is an open standard that creates a bridge between large language models and external tools, APIs, and data sources. Think of it as a universal translator that lets AI agents understand … Continue reading “Connecting AI Agents to Microsoft Fabric with GraphQL and the Model Context Protocol (MCP)”  ( 8 min )

This structured plan helps students: Plans | Microsoft Learn Build foundational knowledge of AI in the cloud. Learn how enterprise-level infrastructure supports responsible, scalable AI deployments. Explore governance and monitoring strategies to ensure security and compliance. And the best part? It’s built using Microsoft’s existing training resources plus some brand-new modules to give you an edge. Your AI Readiness Journey on Azure 🎯 Milestone 1: Getting Started with AI on Azure https://learn.microsoft.com/training/paths/introduction-to-ai-on-azure/Begin with the basics—from machine learning concepts to practical uses of Azure AI services. 🛡️ Milestone 2: Infrastructure Essentials https://learn.microsoft.com/training/paths/manage-iam-for-ai-workloads-on-azure/ https://learn.micro…  ( 20 min )

Global Workshops for All Skill Levels We’re hosting a series of free online workshops to introduce you to MCP—available in multiple languages and programming languages! You’ll get hands-on experience building your first MCP server, guided by friendly experts ready to answer your questions. Register now: https://aka.ms/letslearnmcp Who Should Join? This workshop is built for: Students exploring tech careers Beginner devs eager to learn how AI agents and MCP works Curious coders and seasoned pros alike If you’ve got some code curiosity and a laptop, you’re good to go. Workshop Schedule (English Sessions) DateTech FocusRegistration Link July 9C#Join Here July 15JavaJoin Here July 16PythonJoin Here July 17C# + Visual StudioJoin Here July 21TypeScriptJoin Here Multilingual Sessions We’re…  ( 21 min )

The Power of Connected Agents: Building Multi-Agent Systems Imagine trying to build an AI system that can handle complex workflows like managing support tickets, analyzing data from multiple sources, or providing comprehensive recommendations. Sounds challenging, right? That's where multi-agent systems come in! The Develop a multi-agent solution with Azure AI Foundry Agent Services module introduces you to the concept of connected agents  a game changing approach that allows you to break down complex tasks into specialized roles handled by different AI agents. Why Connected Agents Matter As a student developer, you might wonder why you'd need multiple agents when a single agent can handle many tasks. Here's why this approach is transformative: 1. Simplified Complexity: Instead of building …  ( 29 min )

This week in Model Mondays, we put the spotlight on the AI Toolkit for Visual Studio Code - and explore the tools and workflows that make building generative AI apps and agents easier for developers. Read on for my recap. This post was generated with AI help and human revision & review. To learn more about our motivation and workflows, please refer to this document in our website. About Model Mondays Model Mondays is a weekly series designed to help you grow your Azure AI Foundry Model IQ step by step. Each week includes: 5-Minute Highlights – Quick news and updates about Azure AI models and tools on Monday 15-Minute Spotlight – Deep dive into a key model, protocol, or feature on Monday 30-Minute AMA on Friday – Live Q&A with subject matter experts from the Monday livestream If you're lo…  ( 28 min )

In this example, I demonstrate a Windows-based Function App using PowerShell, with deployment via Azure DevOps (ADO) and a Bicep template. Local development is done in VSCode.   Scenario: Your Function App project resides in a shared repository maintained by a team. Each developer works on a separate branch. Whenever a branch is updated, the Function App is deployed to a slot named after that branch. If the slot doesn't exist, it will be automatically created.   How to use it: Create a Function App You can create a Function App using any method of your choice.   Prepare a corresponding repo in Azure DevOps Set up your repo structure for the Function App source code.   Create Function App code using the VSCode wizard In this example, we use PowerShell and create an anonymous H…  ( 25 min )

Today, you can use Custom Instructions in JetBrains to speed up development while staying aligned with your team’s coding standards and personal preferences. Whether you’re working on a solo project or part of a larger team, adhering to consistent coding standards is essential. Custom Instructions provide a mechanism to reinforce team coding guidelines or embed […] The post Customize GitHub Copilot in JetBrains with Custom Instructions appeared first on Microsoft for Java Developers.  ( 22 min )

Today we are releasing patches that impact the latest version of our self-hosted product, Azure DevOps Server. We strongly encourage and recommend that all customers use the latest, most secure release of Azure DevOps Server. You can download the latest version of the product, Azure DevOps Server 2022.2 from the Azure DevOps Server download page. […] The post July Patches for Azure DevOps Server appeared first on Azure DevOps Blog.  ( 21 min )

SharePoint Embedded is a fully managed, cloud-based, API-only document management system that lets you securely integrate your custom web or mobile apps, whether built on Azure or other clouds, with Microsoft 365 file storage. It’s especially ideal for ISVs building multi-tenant apps because content stays within each customer’s Microsoft 365 tenant.    Design apps that include Microsoft 365 Copilot and agent capabilities, connected Office experiences like Word, and Microsoft Purview compliance and data protection, all within your own user experience. Use built-in retrieval augmented generation (RAG) or bring your own models to create intelligent, secure solutions that reason over your business content, support real-time co-authoring, and scale with granular permissions and storage control.…  ( 39 min )

A Fabric workspace identity is an automatically managed service principal that can be associated with a Fabric workspace. Fabric workspaces with a workspace identity can securely read or write to firewall-enabled Azure Data Lake Storage Gen2 accounts through trusted workspace access for OneLake shortcuts. Fabric items can use the identity when connecting to resources that support Microsoft … Continue reading “Fabric Workspace Identity: Removing Default Contributor Access for Workspace Identity”  ( 7 min )

Azure Storage Actions is a fully managed platform designed to automate data management tasks for Azure Blob Storage and Azure Data Lake Storage. You can use it to perform common data operations on millions of objects across multiple storage accounts without provisioning extra compute capacity and without requiring you to write code. Storage task actions can be used to rehydrate the archived blobs in any tier as required. Please note there is no option to set the rehydration priority and is defaulted to Standard one as of now. Note :- Azure Storage Actions are generally available in the following public regions: https://learn.microsoft.com/en-us/azure/storage-actions/overview#supported-regions Azure Storage Actions is currently in PREVIEW in following reions. Please refer: https://learn.mic…  ( 22 min )

We’ve made significant improvements to how Node.js applications are deployed on Azure App Service — with deployment times now up to 8× faster in some cases.  ( 2 min )

Executive Summary AI agents need identity and trust just like humans. In this article, we demonstrate a zero-trust approach to autonomous AI agents by integrating Identity and access management into an enterprise agentic workflow. Using a hotel booking scenario, we show how each AI agent has identity and issued access token for every action. This ensures no implicit trust between entities – every interaction is authenticated and authorized, following the principle of “never trust, always verify”[1]. The solution marries Asgardeo by WSO2(for identity and access management) with Microsoft’s Azure OpenAI service on Azure AI Foundry and Autogen (for powerful generative AI model and Multiagent orchestration). The result is a working example of autonomous agents that carry digital ID tokens, ena…  ( 59 min )

With all our big announcements at Build, you might think we'd kick back and take a break for a few weeks... but fine tuning never stops! We're wrapping up June with a release of direct preference optimization for the 4.1 family of models, fine tuning available in more regions than ever before, and Responses API support for fine-tuned models.  GPT-4.1, GPT-4.1-mini support Direct Preference Optimization (DPO) 😍 GPT-4.1 and GPT-4.1-mini now support Direct Preference Optimization (DPO). DPO is a finetuning technique that adjusts model weights based on human preferences. You provide a prompt along with two responses: a preferred and non-preferred example; using this data, you can align a fine-tuned model to match your own style, preferences, or safety requirements.   Unlike Reinforcement Lea…  ( 23 min )

That’s the promise behind the new pricing model for Azure AI Content Understanding. We’ve restructured how you pay for document, audio, and video analysis—moving from rigid field-based pricing to a flexible, token-based system that lets you pay only for what you use. Whether you're extracting layout from documents or identifying actions in a video, the new pricing structure delivers up to 60% cost savings for many typical tasks and more control over your spend. Why We’re Moving to Token-Based Pricing Field-based pricing was easy to understand, but it didn’t reflect the real work being done. Some fields are simple. Others require deep reasoning, cross-referencing, and contextual understanding. So we asked: What if pricing scaled with complexity? Enter tokens. Tokens are the atomic units of …  ( 35 min )

After several months in private preview and many bug fixes along the way, we’re excited to announce that Markdown support in large text fields is now generally available! 🎉 🦄 How it works By default, all existing and new work items will continue using the HTML editor for large text fields. However, you now have […] The post Markdown Support Arrives for Work Items appeared first on Azure DevOps Blog.  ( 24 min )

The Cloud Native team at Azure is working to make AI on Kubernetes more cost-effective and approachable for a broader range of users.  The post Optimizing memory usage in large language models fine-tuning with KAITO: Best practices from Phi-3  appeared first on Microsoft Open Source Blog.  ( 15 min )

Introduction Ready to explore the fascinating world of Generative AI using your JavaScript skills? This week’s featured repository, Generative AI for Beginners with JavaScript, is your launchpad into the future of application development. Whether you're just starting out or looking to expand your AI toolbox, this open-source GitHub resource offers a rich, hands-on journey. It includes interactive lessons, quizzes, and even time-travel storytelling featuring historical legends like Leonardo da Vinci and Ada Lovelace. Each chapter combines narrative-driven learning with practical exercises, helping you understand foundational AI concepts and apply them directly in code. It’s immersive, educational, and genuinely fun. What You'll Learn 1. 🧠 Foundations of Generative AI and LLMs Start with t…  ( 23 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP Link part 1 Part 2 - What MCP is, including its architecture and core components Link part 2 Part 3 - A demo of MCP — including how to configure it so anyone can run it Link part 3 Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Part 4 - MCP Development In this section we shall discuss creating an MCP Server - specifically, the Colour MCP server that was covered in the previous section, and used in the demo. MCP SDK To help build MCP servers with minimal effort - there are Software Development Kits (SDKS) for the following languages: C# Java Kotlin Python Ruby Swift TypeScript Links to the different SDKS are at https://modelcontex…  ( 26 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP Part 2 - What MCP is, including its architecture and core components Part 3 - A demo of MCP — including how to configure it so anyone can run it Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Part 4 - MCP Development In this section we shall discuss creating an MCP Server - specifically, the Colour MCP server that was covered in the previous section, and used in the demo. MCP SDK To help build MCP servers with minimal effort - there are Software Development Kits (SDKS) for the following languages: C# Java Kotlin Python Ruby Swift TypeScript Links to the different SDKS are at https://modelcontextprotocol.io/ Colors MCP The source …  ( 26 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP Link part 1 Part 2 - What MCP is, including its architecture and core components Link part 2 Part 3 - A demo of MCP — including how to configure it so anyone can run it Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Link part 4 Part 3 - MCP Demo This section shows the use of several MCP tools - specifically, these will provide knowledge about Colors, Timer, Prices and Azure Resources. The demo will use Visual Studio Code and GitHub Copilot Agent Mode. Demo video Configuration The remainder of this section, explains how to configure and run the demo. Project can be cloned from https://github.com/markharrisonorg/mcpdemo. Alternative…  ( 26 min )

A discussion on Model Context Protocol:Part 1 - Why is there a need for MCP Part 2 - What MCP is, including its architecture and core components Part 3 - A demo of MCP — including how to configure it so anyone can run it Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Part 3 - MCP Demo This section shows the use of several MCP tools - specifically, these will provide knowledge about Colors, Timer, Prices and Azure Resources. The demo will use Visual Studio Code and GitHub Copilot Agent Mode. Demo video Configuration The remainder of this section, explains how to configure and run the demo. Project can be cloned from https://github.com/markharrisonorg/mcpdemo. Alternatively: Create an empty folder and open wi…  ( 25 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP Link part 1 Part 2 - What MCP is, including its architecture and core components Part 3 - A demo of MCP — including how to configure it so anyone can run it  Link part 3 Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Link part 4 Part 2 - What is MCP? The Model Context Protocol (MCP) is an open standard designed to enable intelligent AI applications to seamlessly communicate and interoperate with a wide range of external tools and services. MCP abstracts away integration complexities by defining a unified, extensible communication layer that supports context-aware interactions. https://modelcontextprotocol.io/ Overview of MCP Archit…  ( 21 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP Part 2 - What MCP is, including its architecture and core components Part 3 - A demo of MCP — including how to configure it so anyone can run it Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Part 2 - What is MCP? The Model Context Protocol (MCP) is an open standard designed to enable intelligent AI applications to seamlessly communicate and interoperate with a wide range of external tools and services. MCP abstracts away integration complexities by defining a unified, extensible communication layer that supports context-aware interactions. https://modelcontextprotocol.io/ Overview of MCP Architecture MCP is structured around a modul…  ( 20 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP  Part 2 - What MCP is, including its architecture and core components Link part 2  Part 3 - A demo of MCP — including how to configure it so anyone can run it Link Part 3  Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Link part 4 Part 1 - Why MCP? To understand why the Model Context Protocol (MCP) is needed, lets start by looking at the patterns that modern AI Agents rely on. Common AI Patterns Chat The Chat pattern is a conversational interface between an Agent and an AI Endpoint, typically using natural language, though it can also involve structured inputs or metadata depending on the use case.     The first message sent to an…  ( 31 min )

A discussion on Model Context Protocol: Part 1 - Why is there a need for MCP  Part 2 - What MCP is, including its architecture and core components Part 3 - A demo of MCP — including how to configure it so anyone can run it Part 4 - An example of how to develop an MCP server — a potential starter project for connecting to your own knowledge resources Part 1 - Why MCP? To understand why MCP is needed, lets start by looking at the patterns that modern AI Agents rely on. Common AI Patterns Chat The Chat pattern is a conversational interface between an Agent and an AI Endpoint, typically using natural language, though it can also involve structured inputs or metadata depending on the use case.     The first message sent to an AI endpoint includes a system prompt, which defines the AI's behav…  ( 31 min )

This week in Model Mondays, we focus on Small Language Models (SLMs) and Reasoning — and learn how reasoning models leverage inference-time scaling to execute complex tasks, but how can we use these in resource-constrained devices? Read on for my recap of Mojan Javaheripi's insights on Phi-4 reasoning models that are redefining small language models (SLM) for the agentic era of apps. About Model Mondays Model Mondays is a weekly series designed to help you build your Azure AI Foundry Model IQ step by step. Here's how it works: 5-Minute Highlights – Quick news and updates about Azure AI models and tools on Monday 15-Minute Spotlight – Deep dive into a key model, protocol, or feature on Monday 30-Minute AMA on Friday – Live Q&A with subject matter experts from Monday livestream If you want…  ( 29 min )

This article walks you through an application for Procure-to-Pay anomaly detection using Azure AI Foundry Agent Service. It analyzes purchase invoice images to detect procurement anomalies and compliance issues. The key capabilities of Azure AI Foundry it showcases are: Agent Service that performs a multi-step business process with only natural language instructions. The Application has little to no business logic in the code. Invocation of a variety of automated tool actions required during the multi steps business process. It showcases the recently announced turnkey integration with MCP Server, Azure Logic Apps. In addition, it uses a) visual reasoning of images using gpt-4o models, b) turnkey vector search, and c) reasoning to apply business rules and perform p2p anomaly detection. The…  ( 45 min )

.  ( 16 min )

In Azure Batch, to monitor the node performance like CPU or Disk usage users are required to use Azure monitor.  The Azure Monitor service collects and aggregates metrics and logs from every component of the node. Azure Monitor provides you with a view of availability, performance, and resilience. When you create an Azure Batch pool, you can install any of the following monitoring-related extensions on the compute nodes to collect and analyse data. Previously users have leveraged Batch Insights to get system statistics for Azure Batch account nodes, but it is deprecated now and no longer supported. The Log Analytics agent virtual machine (VM) extension installs the Log Analytics agent on Azure VMs and enrols VMs into an existing Log Analytics workspace. The Log Analytics agent is on a depr…  ( 31 min )

Industry: Energy Location: North America Executive Summary: AI-Driven Multi-Agent Knowledge and IT Support Solution for an Energy Industry Firm A North American energy company sought to modernize its legacy knowledge and IT support chatbot, which was underperforming across key metrics. The existing system, built on static rules and scripts, delivered slow and often inaccurate responses, failing to meet the organization’s standards for employee engagement and operational efficiency. To address this challenge, we proposed and designed a cloud-native, AI-powered multi-agent system hosted on Microsoft Azure. Built on the LangGraph orchestration framework and Azure AI Foundry. This solution integrates advanced ai agent hierarchies, allowing for contextual, domain-specific knowledge retrieval an…  ( 22 min )

1. Introduction 1.1 Evolution from Expert Systems to Modern AI The transition from rule-based expert systems to modern AI represents one of the most significant paradigm shifts in computer science[1]. Where the original 1992 paper by Kiper focused on testing deterministic rule-based systems with clear logical pathways, today's AI systems operate through complex neural architectures that process information in fundamentally different ways[2]. Modern AI systems, particularly deep neural networks and transformer models, exhibit emergent behaviors that cannot be easily traced through simple logical paths[3]. Traditional expert systems operated on explicit if-then rules that could be mapped to logical path graphs, making structural testing relatively straightforward[4]. Contemporary AI systems,…  ( 65 min )

We are excited to announce the public preview of IPv6 outbound support in App Service for Windows! Public preview of outbound IPv6 support for Windows multi-tenant apps is supported on all App Service plan SKUs, Functions Consumption, Functions Elastic Premium, and Logic Apps Standard running Windows. This is the next announcement in our series of IPv6 related feature work on App Service. Public preview of Inbound IPv6 Support on App Service multi-tenant This announcement: IPv6 (dual-stack) non-vnet outbound support (multi-tenant) Backlog - IPv6 vnet outbound support (multi-tenant and App Service Environment v3) Backlog - IPv6 vnet inbound support (App Service Environment v3 - both internal and external) Limitations Linux sites are NOT supported at this time but will be available in the…  ( 22 min )

The Model Context Protocol (MCP) has rapidly become the industry standard for connecting AI agents to a wide range of external tools and services in a consistent way. In a matter of months, this protocol has become a hot topic in developer events and forums and has been implemented by companies large and small. With such rapid change comes the need for training and upskilling to meet the moment! That's why, we're planning a series of virtual training events across different languages (both natural and programming) to introduce you to MCP. ⭐ Register: https://aka.ms/letslearnmcp 👩‍💻 Who Should Join? Whether you're a beginner developer, a university student, or a seasoned tech professional, this workshop was designed with you in mind. At each event, experts will guide you through an exciti…  ( 22 min )

Veeam Data Cloud (VDC) provides cloud-native, SaaS-based data resilience for mission-critical environments including Microsoft 365, Entra ID, Azure, and Salesforce. As enterprise data protection needs to evolve, so does the intelligence behind it. With Veeam Intelligence, VDC is leveraging the power of generative AI that helps users get the most value from VDC services using an advanced AI assistant, with the vision to deliver real-time insights through backup telemetry in the future   To build an assistant delivers accurate, reliable, and safe responses, the VDC team partnered with Azure AI Foundry, Microsoft's AI application development platform which provides built-in responsible AI evaluation and testing  at scale.   From Backup to Breakthrough: Enabling Smart, Safe AI Assistance Veeam…  ( 24 min )

At the //Build conference in May 2025, we announced the public preview of Azure AI Voice Live API (Breakout Session 144). Today we are exciting to share some updates to this API and the latest pricing. Recap: What is the Voice Live API and why does it matter? Voice is the next generation interface between humans and computers. In the era of voice-driven technologies, creating smooth and intuitive speech-based systems has become a priority for developers. The Voice Live API simplifies the process by combining essential voice processing components into a unified interface. Whether you're building conversational agents for customer support, automotive assistants, or educational tools, this API is designed to streamline workflows, reduce latency, and deliver high-quality, real-time voice inter…  ( 38 min )

Copy job has been a go-to tool for simplified data ingestion in Microsoft Fabric, offering a seamless data movement experience from any source to any destination. Whether you need batch or incremental copying, it provides the flexibility to meet diverse data needs while maintaining a simple and intuitive workflow. We continuously refine Copy job based … Continue reading “Simplifying Data Ingestion with Copy job – Incremental Copy GA, Lakehouse Upserts, and New Connectors”  ( 7 min )

The Java ecosystem around Azure Cosmos DB has quietly grown a lot stronger in the first half of 2024. And the updates are especially exciting if you’re working on scalable apps or exploring GenAI workloads. From new support for vector search and indexing, to better throughput management and improved integrations with Apache Kafka, Spring Data, […] The post What’s New for Java Developers in Azure Cosmos DB (NoSQL)? appeared first on Microsoft for Java Developers.  ( 22 min )

Now, with Radius Resource Types, platform engineers can define resource types specific to their organizations. The post Expanding platform engineering capabilities with Radius Resource Types appeared first on Microsoft Open Source Blog.  ( 13 min )

With built-in protections, prevent data leaks, block unsafe prompts, and avoid oversharing without rewriting your app. As a developer, focus on innovation while meeting evolving security and compliance requirements. And as a security admin, gain full visibility into AI data interactions, user activity, and policy enforcement across environments.     Shilpa Ranganathan, Microsoft Purview Principal GPM, shares how new SDKs and Azure AI Foundry integrations bring enterprise-grade security to custom AI apps. Stop data leaks.     Detect and block sensitive content in real-time with Microsoft Purview. Get started. Adapt AI security based on user roles.     Block or allow access without changing your code. See it here. Prevent oversharing with built-in data protections.     Only authorized us…  ( 40 min )

Read how the Retrieval API gives developers a secure, compliant and scalable way to integrate enterprise content into their AI workflows. The post Microsoft 365 Copilot APIs: Unlocking enterprise knowledge for AI with the Retrieval API — Now in Public Preview appeared first on Microsoft 365 Developer Blog.  ( 26 min )

This week in Model Mondays, we focus on the Model Context Protocol (MCP) — and learn how to securely connect AI models to real-world tools and services using MCP, Azure AI Foundry, and industry-standard authorization. Read on for my recap   About Model Mondays Model Mondays is a weekly series designed to help you build your Azure AI Foundry Model IQ step by step. Here’s how it works: 5-Minute Highlights – Quick news and updates about Azure AI models and tools on Monday 15-Minute Spotlight – Deep dive into a key model, protocol, or feature on Monday 30-Minute AMA on Friday – Live Q&A with subject matter experts from Monday livestream If you want to grow your skills with the latest in AI model development, Model Mondays is the place to start. Want to follow along? Register Here - to watc…  ( 31 min )

💡 Why This Approach? Traditional deployments often include models inside the app, leading to: Large container sizes Long build times Slow cold starts Painful updates when models change With Azure Blob Storage, you can offload the model and only fetch it at runtime — reducing size, improving flexibility, and enabling easier updates. What You will Need An ML model (model.pkl, model.pt, etc.) An Azure Blob Storage account A Python web app (FastAPI, Flask, or Streamlit) Azure Web App (App Service for Python) Azure Python SDK: azure-storage-blob Step 1: Save and Upload Your Model to Blob Storage First, save your trained model locally: # PyTorch example import torch torch.save(model.state_dict(), "model.pt") Then, upload it to Azure Blob Storage: from azure.storage.blob import BlobServiceC…  ( 26 min )

Introduction Organizations enforcing Entra ID authentication on their Application Insights resources often face a sudden problem: browser-based telemetry stops flowing. This happens when local authentication is disabled — a necessary step to enforce strict identity controls — but sending data from browser environments comes with inherent security challenges, and the Application Insights JavaScript SDK is no exception. As a result, telemetry from web clients is silently dropped, leaving gaps in monitoring and frustrated teams wondering how to re-enable secure telemetry ingestion. This article provides a solution: using Azure API Management (APIM) as a secure proxy that authenticates telemetry using a managed identity before forwarding it to Application Insights. This pattern restores observ…  ( 34 min )

Anthropic Claude Sonnet 4 and Claude Opus 4 are now available in GitHub Copilot Chat for JetBrains IDEs and Eclipse. Model Availability✨ Claude Sonnet 4: Available for users on Pro, Pro+, Business, and Enterprise plans. Claude Opus 4: Available for users on Pro+ and Enterprise plans. For full details, please see this documentation. How to Get Started🚀 JetBrains IDEs: Click the GitHub Copilot icon –> Open GitHub Copilot Chat-> […] The post Anthropic Claude Sonnet 4 and Opus 4 Now Available in GitHub Copilot for JetBrains and Eclipse appeared first on Microsoft for Java Developers.  ( 23 min )

Shortcut transformations is a new capability in Microsoft Fabric that simplifies the process of converting raw files, starting with .CSV files, into Delta tables. This feature eliminates the need for traditional ETL pipelines, enabling users to transform data directly on top of files with minimal setup. Why use Shortcut transformations Shortcut transformations help users: What … Continue reading “Announcing Shortcut Transformations: from files to Delta tables. Always in sync, no pipelines required.”  ( 7 min )

Introduction and Motivation The explosive growth of data-driven applications has pushed distributed database systems to their limits, especially as organizations demand real-time consistency, high availability, and efficient resource utilization across global infrastructures. The CAP theorem—asserting that a distributed system can guarantee at most two out of consistency, availability, and partition tolerance—forces architects to make challenging trade-offs. Traditional distributed databases rely on static policies and heuristics, which cannot adapt to the dynamic nature of modern workloads and evolving data relationships. Recent advances in Graph Neural Networks (GNNs) offer a new paradigm for modeling and optimizing distributed systems. Unlike conventional machine learning, GNNs naturall…  ( 30 min )

We are delighted to announce the availability of the Voice Conversion (VC) feature in Azure AI Speech service, which is currently in preview. What is voice Conversion Voice Conversion (or voice changer, speech to speech conversion) is the process of transforming the voice characteristics of a given audio to a target voice speaker, and after Voice Conversion, the resulting audio reserves source audio’s linguistic content and prosody while the voice timbre sounds like the target speaker. Below is a diagram of Voice Conversion.   The purpose of Voice Conversion There are 3 reasons users need Voice Conversion functionality: Voice Conversion can replicate your content using a different voice identity while maintaining the original prosody and emotion. For instance, in education, teachers can …  ( 30 min )

Proactively manage and secure all your devices — whether they run Windows, macOS, iOS, or Android. With cross-platform analytics, multi-device queries, and in-depth troubleshooting tools, you can pinpoint problems fast and take targeted action at scale.  Even without deep technical expertise, you can generate complex queries, identify vulnerabilities, and deploy remediations — all in a few clicks. With built-in Copilot support, daily tasks like policy validation, device comparison, and risk analysis become faster, smarter, and easier to act on.  Jeremy Chapman, Director of Microsoft 365, shares how to stay ahead of potential issues and keep endpoints running smoothly.  Spot and fix performance issues before users contact support.  Use Advanced Analytics in Microsoft Intune. Check it out.…  ( 42 min )

The June 2025 release of the on-premises data gateway is (version 3000.274). What’s New Fabric pipeline – Azure database for PostgreSQL connector version 2.0 is now generally available. This new version is enhanced to support TLS 1.3, new table action – upsert, as well as the script activity in data pipeline.  Fabric pipeline – Enhanced … Continue reading “On-premises data gateway June 2025 release”  ( 6 min )

The Microsoft Fabric Extension for VS Code introduces two new features that enhance the management of Fabric items directly within the editor. Users can now perform CRUD operations on Fabric items and switch between multiple tenants easily. These updates aim to improve workflow efficiency and are based on customer feedback, inviting further suggestions for enhancement.  ( 6 min )

Introducing Dev Proxy v0.29, with a major architectural overhaul, control over language model prompts, and improved diagnostics. The post Dev Proxy v0.29 with refactored architecture, MCP server, and exposed LM prompts appeared first on Microsoft 365 Developer Blog.  ( 24 min )

Building robust, scalable AI apps is tough, especially when you want to move fast, follow best practices, and avoid being bogged down by endless setup and configuration. In this quest, you’ll discover how to accelerate your journey from prototype to production by leveraging ready-made templates and modern cloud tools. Say goodbye to decision fatigue and hello to streamlined, industry-approved workflows you can make your own. 👉 Want to catch up on the full program or grab more quests? https://aka.ms/JSAIBuildathon 💬 Got questions or want to hang with other builders? Join us on Discord — head to the #js-ai-build-a-thon channel. 🚀 What You’ll Build A fully functional AI application deployed on Azure, customized to solve a real problem that matters to you.  A codebase powered by a producti…  ( 23 min )

Building robust, scalable AI apps is tough, especially when you want to move fast, follow best practices, and avoid being bogged down by endless setup and configuration. In this quest, you’ll discover how to accelerate your journey from prototype to production by leveraging ready-made templates and modern cloud tools. Say goodbye to decision fatigue and hello to streamlined, industry-approved workflows you can make your own. 👉 Want to catch up on the full program or grab more quests? https://aka.ms/JSAIBuildathon 💬 Got questions or want to hang with other builders? Join us on Discord — head to the #js-ai-build-a-thon channel. 🚀 What You’ll Build A fully functional AI application deployed on Azure, customized to solve a real problem that matters to you.  A codebase powered by a producti…  ( 23 min )

Azure DevOps will no longer depend on the Azure Resource Manager (ARM) resource (https://management.azure.com) when you sign in or refresh Microsoft Entra access tokens. Previously, Azure DevOps required the ARM audience during sign-in and token refresh flows. This requirement meant administrators had to allow all Azure DevOps users to bypass ARM-based Conditional Access policies (CAPs) […] The post Removing Azure Resource Manager reliance on Azure DevOps sign-ins appeared first on Azure DevOps Blog.  ( 23 min )

One of the highly requested features in Microsoft Fabric is now available: the ability to encrypt data in OneLake using your own keys. As organizations face growing data volumes and tighter regulatory expectations, Customer-Managed Keys (CMK) offer a powerful way to enforce enterprise-grade security and ensure strict ownership of encryption keys and access. With Microsoft’s … Continue reading “Customer Managed Keys in OneLake: Strengthening Data Protection and Control”  ( 7 min )

We’re excited to introduce Data Source Instructions, a powerful new feature in the Fabric Data Agent that helps you get more precise, relevant answers from your structured data. What are Data Source instructions? When you use the Data Agent to ask questions in natural language, the agent must determine which data source to use and … Continue reading “New in Fabric Data Agent: Data source instructions for smarter, more accurate AI responses”  ( 6 min )

Welcome to the June 2025 update. The June 2025 Fabric update introduces several key enhancements across multiple areas. Power BI celebrates its 10th anniversary with a range of community events, contests, expert-led sessions, and special certification exam discounts. In Data Engineering, Fabric Notebooks now support integration with variable libraries in preview, empowering users to manage … Continue reading “Fabric June 2025 Feature Summary”  ( 17 min )

Special credit to Rory Preddy for creating the GitHub resource that enable us to learn more about Azure Sora. Reach him out on LinkedIn to say thanks. Introduction Artificial Intelligence (AI) is revolutionizing content creation, and video generation is at the forefront of this transformation. OpenAI's Sora, a groundbreaking text-to-video model, allows creators to generate high-quality videos from simple text prompts. When paired with the powerful infrastructure of Azure AI Foundry, you can harness Sora's capabilities with scalability and efficiency, whether on a local machine or a remote setup.   In this blog post, I’ll walk you through the process of generating AI videos using Sora on Azure AI Foundry. We’ll cover the setup for both local and remote environments. Requirements: Azure AI …  ( 26 min )

We’re excited to announce a significant update to Semantic Kernel Python’s vector store implementation. Version 1.34 brings a complete overhaul that makes working with vector data simpler, more intuitive, and more powerful. This update consolidates the API, improves developer experience, and adds new capabilities that streamline AI development workflows. What Makes This Release Special? The […] The post Semantic Kernel Python Gets a Major Vector Store Upgrade appeared first on Semantic Kernel.  ( 25 min )

Surface the highest-risk alerts across your environment, no matter their default severity, and take action. Customize how your agents reason, teach them what matters to your organization, and continuously refine to reduce time-to-resolution.    Talhah Mir, Microsoft Purview Principal GPM, shows how to triage, investigate, and contain potential data risks before they escalate.  Focus on the most high-risk alerts in your queue.  Save time by letting Alert Triage Agents for DLP and IRM surface what matters. Watch how it works. Stay in control.  Tailor triage priorities with your own rules to focus on what really matters. See how to customize your alert triage agent using Security Copilot. View alert triage agent efficiency stats.  Know what your agent is doing and how well it’s performing…  ( 28 min )

It’s been almost 3 months since we announced OneLake security at FabCon 2025 in Las Vegas, and while the interest has not slowed down, we’ve also been working behind the scenes to improve the feature and address your feedback. In this blog post, we’ll go through some of the latest updates on OneLake security including … Continue reading “OneLake security – updates and news”  ( 6 min )

We’re excited to announce that one of our most-requested features is officially in Public Preview: You can now create Dev Boxes on behalf of your developers. Waiting around to get started is a thing of the past.  Whether you’re onboarding a new hire, running a hackathon, or setting up for a customer demo, this feature makes it […] The post 🎉 Now in Public Preview: Create Dev Boxes on Behalf of Your Developers  appeared first on Develop from the cloud.  ( 23 min )

Authors: Joel Borellis, Mohamad Al Jazaery, Hwasu Kim, Kira Soderstrom  Github Link  This repository is a great starting point. It demonstrates deploying an MCP server with Azure Container Apps and includes three example clients, each using a different agent framework to interact with the server. What’s Inside MCP Server with Sport News ToolsThe sample server, built using the fastmcp package, exposes tools like “Get NFL News”. It supports API key authentication and is designed to be easily extended with additional tools or data sources. You can run it locally or deploy it to Azure Container Apps for scalable, cloud-native hosting.   Three Client SamplesEach example demonstrates how different agent frameworks can consume tools from the MCP server. All the examples use Azure-OpenAI as a…  ( 25 min )

In app-only access scenarios, Shifts Management Graph APIs previously required the MS-APP-ACTS-AS: userId header to indicate the user on whose behalf the application was acting. However, this conflicted with the Microsoft Graph permission model where there is no signed-in user for app-only access scenarios. To align Shifts Graph APIs with this model, the MS-APP-ACTS-AS header […] The post Deprecation of MS-APP-ACTS-AS header in Shifts Management Microsoft Graph APIs appeared first on Microsoft 365 Developer Blog.  ( 24 min )

We’re excited to announce a major enhancement to the deployment experience for Strapi on Azure App Service. Building on the foundation laid out in our  overview, quick start, and FAQ , this update introduces automated and secure secrets management using Azure Key Vault. What’s New? The updated ARM template now provisions an Azure Key Vault instance alongside your Strapi application. This integration enables secure storage of sensitive credentials such as database passwords and Strapi-specific secrets. Here’s what makes this enhancement powerful: Secure by Default: Public access to the Key Vault is disabled out of the box. Instead, private endpoints are configured to ensure secure communication within your virtual network. Auto-Generated Secrets: Strapi secrets are now automatically genera…  ( 24 min )

In our previous blog post Inline Scalar user-defined functions in Microsoft Fabric Warehouse (Preview) we have announced the availability of SQL native scalar UDFs. We also emphasized the importance of inlining and how that can affect scenarios in which UDF can be used. In this post, we aim to highlight common patterns that prevent inlining … Continue reading “How to make your SQL scalar user-defined function (UDF) inlineable in Microsoft Fabric Warehouse “  ( 9 min )

SQL native Scalar user-defined functions (UDFs) in Microsoft Fabric Warehouse and SQL analytics endpoint are now in preview. A scalar UDF is a custom code implemented in T-SQL that accepts parameters, performs an action such as complex calculation, and returns a result of that action as a single value. They can contain local variables, calls … Continue reading “Inline Scalar user-defined functions (UDFs) in Microsoft Fabric Warehouse (Preview)”  ( 8 min )

The JS AI Build-a-thon is in full swing — and we’re turning up the power in Quest 7!If you're just joining us, this is part of an ongoing challenge to help JavaScript and TypeScript developers build AI-powered apps from scratch. Catch up here and join the conversation on Discord.   Last quest, you dipped your toes into agentic development — giving your AI the ability to act and reason.This time, we’re taking it further. In Quest 7, you’ll explore the Model Context Protocol (MCP) — a growing protocol in Agentic development that unlocks standardized tool-usage in AI agents via an MCP server. Available tools for the OS Patrol agent 🎯 What You’ll Build This quest is all about connecting your AI agent to tools that do real things.You’ll: Create and spin up an MCP server using the MCP TypeScri…  ( 25 min )

Let’s face it. Sometimes your Dev Box just… hangs around too long. Whether you’ve moved to a new project, left the company, want to create a new dev box with the latest tools, it’s time to clean things up. 🎯 With Dev Box Auto-Deletion now in public preview, offboarding just got a whole lot easier. […] The post 💻 Spring Cleaning for Dev Boxes: Mastering Manual & Automatic Offboarding appeared first on Develop from the cloud.  ( 24 min )

About Model Mondays Want to know what Reasoning models are and how you can build advanced reasoning scenarios like a Deep Research agent using Azure AI Foundry? Check out this recap from Model Mondays Season 2 Ep 1. Model Mondays is a weekly series to help you build your model IQ in three steps:1. Catch the 5-min Highlights on Monday, to get up to speed on model news2. Catch the 15-min Spotlight on Monday, for a deep-dive into a model or tool3. Catch the 30-min AMA on Friday, for a Q&A session with subject matter experts Want to follow along? Register Here- to watch upcoming livestreams for Season 2 Visit The Forum- to see the full AMA schedule for Season 2 Register Here - to join the AMA on Friday Jun 20 Spotlight On: Advanced Reasoning This week, the Model Mondays spotlight was on Adv…  ( 31 min )

If you're curious about building AI solutions but don’t know where to start, Microsoft’s AI Toolkit is a great place to begin. Whether you’re a student, developer, or just someone exploring AI for the first time, this toolkit helps you build real-world solutions using Microsoft’s powerful AI services. In this blog, I’ll Walk you through what the AI Toolkit is, how you can get started, and where you can find helpful demos and ready-to-use code samples. What is the AI Toolkit? The AI Toolkit is a collection of tools, templates, and sample apps that make it easier to build AI-powered applications and copilots using Microsoft Azure. With the AI Toolkit, you can: Build intelligent apps without needing deep AI expertise. Use templates and guides that show you how everything works. Quickly proto…  ( 24 min )

Table of Contents Abstract Introduction Using Ansys Access with Azure NetApp Files Architecture Diagram Ansys Redhawk Scenario Details Overview and Context HPC Simulation Environment Cloud Shift Drivers Why Azure NetApp Files Capacity and Scale Fits for Ansys Access to support RedHawk-SC Use Cases Power Integrity Simulations Transient Simulations with Frequent Checkpoints Multiple Concurrent Simulation Runs End-to-End Engineering Workflows Azure Well-Architected Pillars And Considerations Performance Efficiency Parallel I/O and Low Latency Large Volumes Dynamic Service Levels and Volume Resizing Protocol Optimization Performance Isolation and QoS Cluster Right-Sizing Cost Optimization Pay-As-You-Go Model Storage Efficiency for data protection Reserved Capacity Tiering for Cold Data Archiva…  ( 41 min )

As organizations scale their development efforts, managing access to cloud resources becomes critical. Platform engineers need to strike a balance between enforcing governance and enabling developer agility. At Build 2025, we announced the general availability of project policies in Microsoft Dev Box, which provides a powerful way to improve resource control and governance for cloud […] The post Balance governance and flexibility with Dev Box project policies appeared first on Develop from the cloud.  ( 24 min )

Cost is one of the most important concerns in any cloud-native rollout. IT admins need powerful tools to control costs without slowing development. At Build 2025, we were excited to announce the general availability of hibernation in Microsoft Dev Box. This feature empowers platform engineers to optimize resource usage while empowering developers to get what […] The post Control cloud costs with Dev Box hibernation features appeared first on Develop from the cloud.  ( 23 min )

Our Team (Sorted Alphabetically): Ashkan Allahyari| ashkan.allahayri@ru.nl Ole Bekker| ole.bekker2@ru.nl Robin Elster| robin.elster@ru.nl Waad Hegazy| waad.hegazy@ru.nl Lea Hierl| lea.hierl@ru.nl Linda Pham| linda.pham@ru.nl  Master Business Analysis & Modelling, Radboud University Master Strategic Human Resources Leadership, Radboud University Student Exchange at Radboud University Project Overview At Radboud University, our team in the course Data-Driven Analytics for Responsible Business Solutions, embraced a unique opportunity to apply our passion for data analytics to a real-world challenge. Tasked with analyzing employee turnover at VenturaGear—a company committed to fostering a thriving workplace—we conducted an in-depth study to uncover the root causes of attrition. By le…  ( 36 min )

What is the Root Cert API? The Root Cert API allows customers to programmatically add root certificates to their ASE, making them available during the startup of apps. Root certificates are public certificates that identify a root certificate authority (CA). These are essential for establishing trust in secure communications. By adding root certificates to your ASE, all web apps hosted within that ASE will have them installed in their root store. This ensures that apps can securely communicate with internal services or APIs that use certificates issued by private or enterprise CAs. Previously, this functionality was only available in private preview through a workaround involving certificate uploads and a special app setting and included a number of limitations. With the new Root Cert API,…  ( 27 min )

The Model Context Protocol (MCP) is rapidly becoming the prominent framework for building truly agentic, interoperable AI applications.   Many articles document MCP servers for single-server use, this project stands out as the starter template that combines Azure OpenAI integration with a Multi-Server MCP architecture on a custom interface, enabling you to connect and orchestrate multiple tool servers—through a customized UI.   Here, we will deep dive into Multi Server MCP implementation, connecting both local custom and ready MCP Servers in a single client session through the MultiServerMCP library from Langchain adapters, enabling agentic orchestration across different domains. While most triggers to OOB MCP servers leveraged Github Copilot for input, this project allows for custom app i…  ( 31 min )

In 2022, we announced the deprecation of LUIS by September 30, 2025 with a recommendation to migrate to conversational language understanding (CLU). In response to feedback from our valued customers, we have decided to extend the availability of certain functionalities in LUIS until March 31, 2026. This extension aims to support our customers in their smooth migration to CLU, ensuring minimal disruption to their operations.  Extension Details  Here are some details on when and how the LUIS functionality will change:  October 2022: LUIS resource creation is no longer available. October 31, 2025: The LUIS portal will no longer be available.  LUIS Authoring (via REST API only) will continue to be available. March 31, 2026: LUIS Authoring, including via REST API, will no longer be available.  LUIS Runtime will no longer be available.  Before these retirement dates, please migrate to conversational language understanding (CLU), a capability of Azure AI Service for Language. CLU provides many of the same capabilities as LUIS, plus enhancements such as:  Enhanced AI quality using state-of-the-art machine learning models  The LLM-powered Quick Deploy feature to deploy a CLU model with no training Multilingual capabilities that allow you to train in one language and predict in 99+ others  Built-in routing between conversational language understanding and custom question answering projects using orchestration workflow  Access to a suite of features available on Azure AI Service for Language in the Azure AI Foundry  Looking Ahead  On March 31, 2026, LUIS will be fully deprecated, and any LUIS inferencing requests will return an error message. We encourage all our customers to complete their migration to CLU as soon as possible to avoid any disruptions.  We appreciate your understanding and cooperation as we work together to ensure a smooth migration.   Thank you for your continued support and trust in our services.  ( 21 min )

Introduction As sustainability becomes a core priority for organizations worldwide, Azure Migrate now empowers customers to quantify environmental impact alongside cost savings when planning their cloud journey. With the new Sustainability Benefits capability in Azure Migrate's Business Case, customers can now view estimated emissions savings when migrating from on-premises environments to Azure — making sustainability a first-class consideration in cloud transformation. Align with Global Sustainability Goals With governments and enterprises racing to meet net-zero targets — including a 55% emissions reduction target by 2030 in the EU and net-zero goals in the US and UK by 2050 — cloud migration offers a meaningful path to emissions reduction. With Azure’s carbon-efficient infrastructure p…  ( 25 min )

What is Azure AI Foundry and Foundry Local? Azure AI Foundry is Microsoft’s comprehensive platform for enterprise AI development and deployment, enabling organizations to build, customize, and operate AI solutions at scale. It provides tools, services, and infrastructure to develop, fine-tune and deploy AI models in production environments with enterprise-grade security and compliance. Foundry Local […] The post Connect Spring AI to Local AI Models with Foundry Local appeared first on Microsoft for Java Developers.  ( 25 min )

A few weeks ago at BUILD, we announced the upcoming Azure DevOps MCP Server: 👉 Azure DevOps with GitHub Repositories – Your path to Agentic AI Today, we’re excited to share that the local Azure DevOps MCP Server is now available in public preview. This lets GitHub Copilot in Visual Studio and Visual Studio Code […] The post Azure DevOps MCP Server, Public Preview appeared first on Azure DevOps Blog.  ( 23 min )

Promoting an application or infrastructure change into production often comes with a requirement to follow a change control process. This ensures that changes to production are properly reviewed and that they adhere to required approvals, change windows and QA process. Often this change request (CR) process will be conducted using a system for recording and auditing the change request and the outcome. When deploying a release, there will often be places in the process to go through this change control workflow. This may be as part of a release pipeline, it may be managed in a pull request or it may be a manual process. Ultimately, by the time the actual changes are made to production infrastructure or applications, they should already be approved. This relies on the appropriate controls an…  ( 40 min )

TLDR Create an Azure Kubernetes Service cluster with GPU nodes and connect it to Azure Machine Learning to run distributed ML training workloads. This integration provides a managed data science platform while maintaining Kubernetes flexibility under the hood, enables multi-node training that spans multiple GPUs, and bridges the gap between infrastructure and ML teams. The solution works for both new and existing clusters, supporting specialized GPU hardware and hybrid scenarios. Why Should You Care? Integrating Azure Kubernetes Service (AKS) clusters with GPUs into Azure Machine Learning (AML) offers several key benefits: Utilize existing infrastructure: Leverage your existing AKS clusters with GPUs via a managed data science platform like AML Flexible resource sharing: Allow both AKS wo…  ( 34 min )

Co-author: Alexei Robsky, Data Scientist Manager Overview  As organizations increasingly rely on real-time data to drive decisions, the need for intelligent, responsive systems has never been greater. At the heart of this transformation is Fabric Real-Time Intelligence (RTI), a platform that empowers users to act on data as it arrives. Today, we’re excited to announce … Continue reading “Introducing MCP Support for Real-Time Intelligence (RTI) “  ( 7 min )

The Eventstreams artifact in the Microsoft Fabric Real-Time Intelligence experience lets you bring real-time events into Fabric, transform them, and then route them to various destinations such as Eventhouse, without writing any code (no-code). You can ingest data from an Eventstream to Eventhouse seamlessly either from Eventstream artifact or Eventhouse Get Data Wizard. This capability … Continue reading “Fabric Eventhouse now supports Eventstream Derived Streams in Direct Ingestion mode (Preview)”  ( 7 min )

Have you ever found yourself frustrated by inconsistent item creation? Maybe you’ve struggled to select the right workspace or folder when creating a new item or ended up with a cluttered workspace due to accidental item creation. We hear you—and we’re excited to introduce the new item creation experience in Fabric! This update is designed … Continue reading “Introducing new item creation experience in Fabric”  ( 6 min )

We’re excited to announce Surge Protection for background operations is now Generally Available (GA). Using surge protection, capacity admins can limit overuse by background operations in their capacities.  ( 6 min )

Scenario Once a loan application is submitted, financial institutions must process a variety of supporting documents—including pay stubs, tax returns, credit reports, and bank statements—before a loan can be approved. This post-application phase is often fragmented and manual, involving data retrieval from multiple systems, document verification, eligibility calculations, packet compilation, and signing. Each step typically requires coordination between underwriters, compliance teams, and loan processors, which can stretch the processing time to several weeks. This solution automates the post-application loan processing workflow using Azure services and Generative AI agents. Intelligent agents retrieve and validate applicant data, extract and summarize document contents, calculate loan eli…  ( 42 min )

The June 2025 release includes Copilot chat tools in the Python extension, project creation from a template, language server based terminal suggest, and more! The post Python in Visual Studio Code – June 2025 Release appeared first on Microsoft for Python Developers Blog.  ( 24 min )

Result Set Caching is now available in preview for Microsoft Fabric Data Warehouse and Lakehouse SQL analytics endpoint. This performance optimization works transparently to cache the results of eligible T-SQL queries. When the same query is issued again, it directly retrieves the stored result, instead of recompiling and recomputing the original query. This operation drastically … Continue reading “Result Set Caching for Microsoft Fabric Data Warehouse (Preview)”  ( 6 min )

In this quest, you'll teach your AI app to talk to external data using the Retreival Augmented Generation (RAG) technique. You'll overcome the limitations of pre-trained language models by allowing them to reference your own data, using it as context to deliver accurate, fact-based responses.   👉 Want to catch up on the full program or grab more quests? https://aka.ms/JSAIBuildathon 💬 Got questions or want to hang with other builders? Join us on Discord — head to the #js-ai-build-a-thon channel. 🔧 What You’ll Build In this quest, you’ll: Connect your AI app to external documents (like PDFs) Allow your app to “read” and respond using your real-world content Why does this matter? Because LLMs are powerful, but they don’t know your business, reports, or research papers, etc. With RAG, yo…  ( 25 min )

Prompty is a powerful tool for managing prompts in AI applications. Not only does it allow you to easily test your prompts during development, but it also provides observability, understandability and portability. Here's how to use Prompty with Foundry Local to support your AI applications with on-device inference. Foundry Local At the Build '25 conference, Microsoft announced Foundry Local, a new tool that allows developers to run AI models locally on their devices. Foundry Local offers developers several benefits, including performance, privacy, and cost savings. Why Prompty? When you build AI applications with Foundry Local, but also other language model hosts, consider using Prompty to manage your prompts. With Prompty, you store your prompts in separate files, making it easy to test a…  ( 27 min )

The Fashion Trends Discovery Scenario In this walkthrough, we'll explore a sample application that demonstrates the power of combining Computer Use (CUA) models with Playwright browser automation to autonomously compile trend information from the internet, while leveraging MCP integration to intelligently catalog and store insights in Azure Blob Storage. The User Experience A fashion analyst simply provides a query like "latest trends in sustainable fashion" to our command-line interface. What happens next showcases the power of agentic AI—the system requires no further human intervention to: Autonomous Web Navigation: The agent launches Pinterest, intelligently locates search interfaces, and performs targeted queries Intelligent Content Discovery: Systematically identifies and interacts …  ( 43 min )

Let me walk you through 12 features that have become indispensable in my daily coding life. Understanding Code Changes & History Inline Blame Annotations: GitLens adds small annotations at the end of each line of code, showing who last modified that line, when, and in which commit. This feature provides instant context about the code's history without leaving your editor. While working on an e-commerce platform's checkout process, I noticed an unexpected behavior. With inline blame, I quickly identified that the change was introduced in a recent sprint, saving hours of backtracking.  Heatmap: This feature adds a color-coded heatmap to the scroll bar, visually representing the age of the code. Newer changes appear in warmer colors, while older code is shown in cooler colors, helping you q…  ( 30 min )

In this quest, you’ll give your Gen AI prototype a polished chat interface using Vite and Lit. Along the way, you’ll also manage application infrastructure with Bicep and Azure Developer CLI (azd), making your prototype more structured and ready for deployment. This step is all about UX, making your AI prototype not just functional, but interactive and user-friendly. 👉 Want to catch up on the full program or grab more quests? https://aka.ms/JSAIBuildathon 💬 Got questions or want to hang with other builders? Join us on Discord — head to the #js-ai-build-a-thon channel. 🔧 What You’ll Build By the end of this quest, you’ll have: A chat UI built with Vite and Lit  A structured codebase with infrastructure-as-code (IaC) using Bicep  Seamless local deployment workflow using the Azure Develop…  ( 26 min )

Over the course of the last year, we have launched several Cohere models on Azure as Serverless Standard (pay-go) offering. We’re excited to announce that Cohere's latest models—Command A, Rerank 3.5, and Embed 4—are now available on Azure AI Foundry models via Managed Compute.   This launch allows enterprises and developers to now deploy Cohere models instantly with their own Azure quota, with per-hour GPU pricing that compensates the model provider—unlocking a scalable, low-friction path to production-ready GenAI.    What is Managed Compute?  Managed Compute is a deployment option within Azure AI Foundry Models that lets you run large language models (LLMs), SLMs, HuggingFace models and custom models fully hosted on Azure infrastructure.    Why Use Managed Compute?  Azure Managed Compute…  ( 23 min )

Introduction Quarkus is a general-purpose Java framework focused on efficient use of resources, fast startup, and rapid development. It allows developers to create and run services in the Java Virtual Machine (JVM) or native binary executables (native mode). In this blog post we are going to focus on using Quarkus to create and monitor a […] The post Monitor your Quarkus native application on Azure appeared first on Microsoft for Java Developers.  ( 25 min )

This post goes over situations where you may notice jobs being terminated or stopped briefly and some things that can be done to help alleviate interruptions.  ( 7 min )

Introduction Integrating Azure API Management (APIM) with Microsoft Fabric’s API for GraphQL can significantly enhance your API’s capabilities by providing robust scalability and security features such as identity management, rate limiting, and caching. This post will guide you through the process of setting up and configuring these features. You may not be familiar with API … Continue reading “Integrating Azure API Management with Fabric API for GraphQL”  ( 8 min )

Introduction Whether you’re building analytics pipelines or conversational AI systems, the risk of exposing sensitive data is real. AI models trained on unfiltered datasets can inadvertently memorize and regurgitate PII, leading to compliance violations and reputational damage. This blog explores how to build scalable, secure, and compliant data workflows using PySpark, Microsoft Presidio, and Faker—covering … Continue reading “Privacy by Design: PII Detection and Anonymization with PySpark on Microsoft Fabric”  ( 9 min )

📍 Milano, Sala Luigiana @ Coperni.co Centrale - Via Copernico 38📅 17 Giugno 2025 dalle 9:00 alle 17:30🧠 Focus: AI Agents per sviluppatori, ricercatori, imprenditori e professionisti IT La serie globale AgentCon – AI Agents World Tour arriva a Milano per una giornata interamente dedicata a chi costruisce il futuro con l’intelligenza artificiale. Dopo il successo delle tappe internazionali in Kenya, Olanda e Brasile, il tour approda in Italia per offrire contenuti tecnici, demo dal vivo e networking con esperti del settore. Perché partecipare? AgentCon non è una conferenza qualsiasi. È il luogo dove si parla di casi d'uso concreti e esempi pratici di codice. L’evento è pensato per sviluppatori, architetti software, data scientist e professionisti IT che vogliono andare oltre i modelli e s…  ( 22 min )

Earlier this year, we released AI functions in public preview, allowing Fabric customers to apply LLM-powered transformations to OneLake data simply and seamlessly, in a single line of code. Since then, we’ve continued iterating on AI functions in response to your feedback. Let’s explore the latest updates, which make AI functions more powerful, more cost-effective, … Continue reading “Introducing upgrades to AI functions for better performance—and lower costs”  ( 7 min )

Introduction Let’s take a closer look at an exciting development in the AI space. Embedding models are the key to transforming complex data into usable insights, driving innovations like smarter chatbots and tailored recommendations. With Azure AI Foundry, Microsoft’s powerful platform, you’ve got the tools to build and scale these models effortlessly. Add in Open Web UI, a intuitive interface for engaging with AI systems, and you’ve got a winning combo that’s hard to beat. In this article, we’ll explore how embedding models on Azure AI Foundry, paired with Open Web UI, are paving the way for accessible and impactful AI solutions for developers and businesses. Let’s dive in!   To proceed with configuring the embedding model from Azure AI Foundry on Open Web UI, please firstly configure the…  ( 24 min )

Today we are releasing patches that impact the latest version of our self-hosted product, Azure DevOps Server. We strongly encourage and recommend that all customers use the latest, most secure release of Azure DevOps Server. You can download the latest version of the product, Azure DevOps Server 2022.2 from the Azure DevOps Server download page. […] The post June Patches for Azure DevOps Server appeared first on Azure DevOps Blog.  ( 22 min )

Enforce MFA, block legacy authentication, and apply risk-based Conditional Access policies to reduce exposure from stale accounts and weak authentication methods. Use built-in tools for user, group, and device administration to detect and clean up identity sprawl — like unused credentials, inactive accounts, and expired apps — before they become vulnerabilities.  Jeremy Chapman, Microsoft 365 Director, shares steps to clean up your directory, strengthen authentication, and improve overall identity security.  Prioritize top risks.  Take action across MFA, risk policies, and stale objects with Microsoft Entra recommendations. Start here. Block over 99% of identity attacks.  Enforce MFA for admins and users in Microsoft Entra. Detect and delete stale user accounts.  See how to fix account …  ( 45 min )

The Azure Incubations team is proud to share that Drasi has officially been accepted into the Cloud Native Computing Foundation Sandbox. The post Drasi accepted into CNCF sandbox for change-driven solutions  appeared first on Microsoft Open Source Blog.  ( 13 min )

With Azure Container Apps we can run containerized applications, completely serverless. The platform itself handles all the orchestration needed to dynamically scale based on your set triggers (such as KEDA) and even scale-to-zero! I have been working a lot with customers recently on using Azure API Management (APIM) and the topic of how we can leverage Azure APIM to manage our internal APIs without having to expose a public IP and stay within compliance from a security standpoint, which leads to the use of a Self-Hosted Gateway. This offers a managed gateway deployed within their network, allowing a unified approach in managing their APIs while keeping all API communication in-network. The self-hosted gateway is deployed as a container and in this article, we will go through how to provis…  ( 27 min )

Migrating applications to the cloud or modernizing infrastructure requires thorough preparation. Whether it’s a cloud platform, a new data center, or a hybrid infrastructure — is a complex process. While organizations focus on optimizing performance, costs, and scalability, security often takes a backseat, leading to potential risks post-migration. One crucial step before migration is conducting a pre-migration scan to identify security vulnerabilities, licensing risks, and code quality issues. Several tools help in pre-migration scanning, including Blackduck, Coverity, Gitleaks, and Semgrep. In this article, we will explore the role of these tools in migration readiness. Why Perform a Pre-Migration Scan? When an application moves from an on-premises environment to the cloud, it interacts…  ( 31 min )

We're thrilled to announce the Public Preview of Azure Migrate Explore with an AI assistant! This exciting update enhances our existing Azure Migrate Explore (AME) utility, making migration assessments smarter, faster, and more impactful. What is Azure Migrate? Azure Migrate serves as a comprehensive hub designed to simplify the migration journey of on-premises infrastructure, including servers, databases, and web applications, to Azure Platform-as-a-Service (PaaS) and Infrastructure-as-a-Service (IaaS) targets at scale. It provides a unified platform with a suite of tools and features to help you identify the best migration path, assess Azure readiness, estimate the cost of hosting workloads on Azure, and execute the migration with minimal downtime and risk. Revolutionizing Executive Pres…  ( 22 min )

Microsoft and F5 are collaborating on Phase 2 of the OpenTelemetry with Apache Arrow project. The post Microsoft and F5 join forces on OpenTelemetry with Apache Arrow in Rust appeared first on Microsoft Open Source Blog.  ( 12 min )

Add deep AI integration with built-in vector search and DiskANN optimizations, plus native support for large object JSON and new Change Event Streaming for live data updates.  Join and analyze data faster with the Lakehouse shortcuts in Microsoft Fabric that unify multiple databases — across different SQL Server versions, clouds, and on-prem — into a single, logical schema without moving data. Build intelligent apps, automate workflows, and unlock rich insights with Copilot and the unified Microsoft data platform, including seamless Microsoft Fabric integration, all while leveraging your existing SQL skills and infrastructure.  Bob Ward, lead SQL engineer, joins Jeremy Chapman to share how the latest SQL Server 2025 innovations simplify building complex, high-performance workloads with les…  ( 57 min )

We’re excited to announce that the long-awaited refresh SQL analytics endpoint metadata REST API is now available in preview. You can now programmatically trigger a refresh of your SQL analytics endpoint to keep tables in sync with any changes made in the parent artifact, ensuring that you can keep your data up to date as needed. … Continue reading “Refresh SQL analytics endpoint Metadata REST API (Preview)”  ( 6 min )

Debugging your code is a big deal, especially when you’re working with user data functions. You want to make sure everything works as it should and that’s where local debugging lets you catch problems in your code without messing with the live environment. In this blog post, I’ll walk you through the steps to make local debugging easier, faster, and less of a headache.  ( 7 min )

Artificial Intelligence is evolving, and so are the ways we build intelligent agents. On a recent Microsoft YouTube Live session, developers and AI enthusiasts gathered to explore the power of custom tools in AI agents using Azure AI Studio. The session walked through concepts, use cases, and a live demo that showed how integrating custom tools can bring a new level of intelligence and adaptability to your applications. 🎥 Watch the full session here:  https://www.youtube.com/live/MRpExvcdxGs?si=X03wsQxQkkshEkOT What Are AI Agents with Custom Tools? AI agents are essentially smart workflows that can reason, plan, and act — powered by large language models (LLMs). While built-in tools like search, calculator, or web APIs are helpful, custom tools allow developers to tailor agents for busin…  ( 23 min )

Part of the JS AI Build-a-thon, this quest is where the rubber meets the road. Whether you're here to explore the raw power of local models or just curious what Generative AI looks like under the hood, this guide is for you. 👉 Want to catch up on the full program or grab more quests? Start here💬 Got questions or just want to hang with other builders? Join us on Discord — head to the #js-ai-build-a-thon channel. 🧩 What You’ll Build in This Quest   In this quest, you’ll build a fully working local Gen AI prototype — no cloud APIs, no credits needed. You’ll explore the GitHub Models catalog, run local inference using an open model, and convert a hand-drawn UI sketch into a working webpage. Along the way, you’ll get hands-on with AI developer tooling built right into Visual Studio Code. Wh…  ( 31 min )

Part of the JS AI Build-a-thon, this quest is where the rubber meets the road. Whether you're here to explore the raw power of local models or just curious what Generative AI looks like under the hood, this guide is for you. 👉 Want to catch up on the full program or grab more quests? Start here💬 Got questions or just want to hang with other builders? Join us on Discord — head to the #js-ai-build-a-thon channel. 🧩 What You’ll Build in This Quest   In this quest, you’ll build a fully working local Gen AI prototype — no cloud APIs, no credits needed. You’ll explore the GitHub Models catalog, run local inference using an open model, and convert a hand-drawn UI sketch into a working webpage. Along the way, you’ll get hands-on with AI developer tooling built right into Visual Studio Code. Wh…  ( 31 min )

Want to build next-gen apps using AI and JavaScript (or TypeScript)? The JS AI Build-a-thon is your launchpad — a hands-on, quest-driven journey designed to take you from curious coder to AI-powered app builder. It’s self-paced, community-backed, and packed with real-world use cases. No fluff. Just code, context, and creative exploration. 👾 Ready to roll? Here's your setup guide: 🔧 Step 1: Start Your Journey Head to aka.ms/JSAIBuild-a-thon and find the GitHub repository. Before you hit “Start Course,” take a moment to scroll through the README — it’s packed with useful info, including a global list of Study Jams where you can connect with other developers learning alongside you. Find one near you or join virtually to level up with the community. Screenshot off available study jams Once y…  ( 25 min )

Overview Azure VMware Solution is a VMware validated first party Azure service from Microsoft that provides private clouds containing VMware vSphere clusters built from dedicated bare-metal Azure infrastructure. It enables customers to leverage their existing investments in VMware skills and tools, allowing them to focus on developing and running their VMware-based workloads on Azure. VMware HCX is the mobility and migration software used by the Azure VMware Solution to connect remote VMware vSphere environments to the Azure VMware Solution. These remote VMware vSphere environments can be on-premises, co-location or cloud-based instances.     Figure 1 – Azure VMware Solution with VMware HCX Service Mesh Broadcom has announced the end-of-life (EOL) for VMware HCX version 4.10.x, effective …  ( 27 min )

We are pleased to announce that Azure VMware Solution is now available in Korea Central. Now in 34 Azure regions, Azure VMware Solution empowers you to seamlessly extend or migrate existing VMware workloads to Azure without the cost, effort or risk of re-architecting applications or retooling operations.  Azure VMware Solution supports: Rapid cloud migration of VMware-based workloads to Azure without refactoring. Datacenter exit while maintaining operational consistency for the VMware environment. Business continuity and disaster recovery for on-premises VMware environments. Attach Azure services and innovate applications at your own pace. Includes the VMware technology stack and lets you leverage existing Microsoft licenses for Windows Server and SQL Server. For updates on current and upcoming region availability, visit the product by region page here.   Streamline migration with new offers and licensing benefits, including a 20% discount. We recently announced the VMware Rapid Migration Plan, where Microsoft provides a comprehensive set of licensing benefits and programs to give you price protection and savings as you migrate to Azure VMware Solution. Azure VMware Solution is a great first step to the cloud for VMware customers, and this plan can help you get there. Learn More  ( 19 min )

Use Cases: From Automation to Edge AI   Generative AI is transforming industries through: Content creation, summarization, and translation Customer engagement via chatbots and personalization Edge deployment for low-latency, privacy-sensitive applications Domain-specific tasks like legal, medical, or technical document processing LLMs vs. SLMs: Choosing the Right Fit FeatureLLMsSLMs ParametersBillions (e.g., GPT-4)Millions PerformanceHigh accuracy, nuanced understandingFast, efficient for simpler tasks DeploymentCloud-based, resource-intensiveIdeal for edge and mobile CostHigh compute and energyCost-effective SLMs are increasingly viable thanks to optimized runtimes and hardware, making them perfect for on-device AI. Azure AI Foundry: Your AI Launchpad Azure AI Foundry offers: A model catalogue with open-source and proprietary models   Tools for fine-tuning, evaluation, and deployment Integration with GitHub, VS Code, and Azure DevOps Support for both cloud and local inferencing Local AI: The Edge Advantage With tools like Foundry Local and Windows AI Foundry, developers can: Run models on-device with ONNX Runtime Use APIs for summarization, translation, and more Optimize for CPU, GPU, and NPU Ensure privacy, low latency, and offline capability Customization: RAG vs. Fine-Tuning FeatureRAGFine-Tuning Knowledge UpdatesDynamicStatic InterpretabilityHighLow LatencyHigherLower Hallucination RiskLowerModerate Use CaseReal-time, external dataDomain-specific tasks Both methods enhance model relevance RAG by retrieving external data, and fine-tuning by adapting model weights. Developer Resources Get started with: Foundry Local SDK Windows AI Foundry | Microsoft Developer AI Toolkit for VS Code Windows ML Azure AI Learn Courses Join the Azure AI Discord Community  ( 21 min )

As organizations continue to strengthen their security posture, restricting usage of personal access tokens (PATs) has become a critical area of focus. With the latest public preview of the Restrict personal access token creation policy in Azure DevOps, Project Collection Administrators (PCAs) now have another powerful tool to reduce unnecessary PAT usage and enforce tighter […] The post Restricting PAT Creation in Azure DevOps Is Now in Preview appeared first on Azure DevOps Blog.  ( 25 min )

Orchestrate multi-agent apps and high-scale inference solutions using open-source and proprietary models, no infrastructure management needed. With Azure, connect frameworks like Semantic Kernel to models from DeepSeek, Llama, OpenAI’s GPT-4o, and Sora, without provisioning GPUs or writing complex scheduling logic. Just submit your prompt and assets, and the models do the rest. Using Azure’s Model as a Service, access cutting-edge models, including brand-new releases like DeepSeek R1 and Sora, as managed APIs with autoscaling and built-in security. Whether you’re handling bursts of demand, fine-tuning models, or provisioning compute, Azure provides the capacity, efficiency, and flexibility you need. With industry-leading AI silicon, including H100s, GB200s, and advanced cooling, your solut…  ( 57 min )

In the world of software development, flexibility and adaptability are key. Developers often face challenges when it comes to updating plugin metadata dynamically without disrupting services or requiring redeployment. This is where SemanticPluginForge, an open-source project, steps in to improve the way we manage plugin metadata. LLM Function Calling Feature The function calling feature in LLMs […] The post Enhancing Plugin Metadata Management with SemanticPluginForge appeared first on Semantic Kernel.  ( 25 min )

Smarter SK Agents with Contextual Function Selection In today’s fast-paced AI landscape, developers are constantly seeking ways to make AI interactions more efficient and relevant. The new Contextual Function Selection feature in the Semantic Kernel Agent Framework is here to address this need. By dynamically selecting and advertising only the most relevant functions based on […] The post Smarter SK Agents with Contextual Function Selection appeared first on Semantic Kernel.  ( 24 min )

As data volume and complexity soar, protecting sensitive information has become non-negotiable. With the latest enhancements to Purview Data Loss Prevention (DLP) Policies in Microsoft Fabric, organizations now have the power to proactively secure their data in Onelake. Whether you’re just getting started or looking to take your data governance to the next level, following … Continue reading “Secure Your Data from Day One: Best Practices for Success with Purview Data Loss Prevention (DLP) Policies in Microsoft Fabric”  ( 7 min )

Quick Links To RSVP for each episode: EP1: Advanced Reasoning Models: https://developer.microsoft.com/en-us/reactor/events/25905/  EP2: Model Context Protocol:https://developer.microsoft.com/en-us/reactor/events/25906/  EP3: SLMs (and Reasoning):https://developer.microsoft.com/en-us/reactor/events/25907/  Get All The Details: https://aka.ms/model-mondays    Azure AI Foundry offers the best model choice  Did you manage to catch up on all the talks from Microsoft Build 2025? If, like me, you are interested in building AI-driven applications on Azure, you probably started by looking at what’s new in Azure AI Foundry.  I recommend you read Asha Sharma’s post for the top 10 things you need to know in this context. And it starts with New Models & Smarter Models!  New Models | Azure AI Foundr…  ( 28 min )

The latest version of Dev Proxy introduces a new ability to help you understand language models’ usage and costs in your applications, alongside many improvements to mocking, TypeSpec generation, and plugin flexibility. The post Dev Proxy v0.28 with LLM usage and costs tracking appeared first on Microsoft 365 Developer Blog.  ( 25 min )

Why should I care?   ⚡️In under 15 minutes, you’ll have a Cypher-powered, semantically rich knowledge graph you can query interactively.   How can graphRAG help? GraphRAG extracts structured knowledge from raw, unstructured data like .txt files by building a knowledge graph. This enables more precise and context-aware retrieval, making it easier to surface relevant insights from messy or disconnected content. What are the challenges? While the standard graphRAG indexing process typically expects input and output directories, some users already store their data in a DB (database) and prefer to run graphRAG directly against using the DB for both input and output. This eliminates the need for intermediate blob storage and simplifies the pipeline. Additionally, customers often request support …  ( 48 min )

Why should I care?   ⚡️In under 15 minutes, you’ll have a Cypher-powered, semantically rich knowledge graph you can query interactively.   How can graphRAG help? GraphRAG extracts structured knowledge from raw, unstructured data like .txt files by building a knowledge graph. This enables more precise and context-aware retrieval, making it easier to surface relevant insights from messy or disconnected content. What are the challenges? While the standard graphRAG indexing process typically expects input and output directories, some users already store their data in a DB (database) and prefer to run graphRAG directly against using the DB for both input and output. This eliminates the need for intermediate blob storage and simplifies the pipeline. Additionally, customers often request support …

If you haven't seen my previous blog post introducing MCP on Azure App Service, check that out here for a quick overview and getting started. In this blog post, I’m excited to share some updates for our App Service MCP samples: new language samples, updated functionality to replace deprecated methods, and built-in authentication and authorization—all designed to make it easier for developers to host MCP servers on Azure App Service. 🔄 Migrating from SSE to Streamable HTTP The original .NET sample I shared used Server-Sent Events (SSE) for streaming responses. However, SSE has since been deprecated in favor of streamable HTTP, which offers better compatibility and performance across platforms. To align with the latest MCP specification, I’ve updated the .NET sample to use streamable HTTP: …  ( 23 min )

If you haven't seen my previous blog post introducing MCP on Azure App Service, check that out here for a quick overview and getting started. In this blog post, I’m excited to share some updates for our App Service MCP samples: new language samples, updated functionality to replace deprecated methods, and built-in authentication and authorization—all designed to make it easier for developers to host MCP servers on Azure App Service. 🔄 Migrating from SSE to Streamable HTTP The original .NET sample I shared used Server-Sent Events (SSE) for streaming responses. However, SSE has since been deprecated in favor of streamable HTTP, which offers better compatibility and performance across platforms. To align with the latest MCP specification, I’ve updated the .NET sample to use streamable HTTP: …

Take a deep dive into Athena, an AI-powered collaborative agent, to learn how it was built and how to create your own version of Athena right within Microsoft Teams. The post Build like Microsoft: Developer agents in action appeared first on Microsoft 365 Developer Blog.  ( 23 min )

Optimizing JVM Configuration for Azure Deployments Tuning the Java Virtual Machine (JVM) for cloud deployments is notoriously challenging. Over 30% of developers tend to deploy Java workloads with no JVM configuration at all, therefore relying on the default settings of the HotSpot JVM.  The default settings in OpenJDK are intentionally conservative, designed to work across a wide range of environments and scenarios. However, these defaults often lead to suboptimal resource utilization in cloud-based deployments, where memory and CPU tend to be dedicated for application workloads (use of containers and VMs) but still require intelligent management to maximize efficiency and cost-effectiveness. To address this, we are excited to introduce jaz, a new JVM launcher optimized specifically for …  ( 26 min )

Optimizing JVM Configuration for Azure Deployments Tuning the Java Virtual Machine (JVM) for cloud deployments is notoriously challenging. Over 30% of developers tend to deploy Java workloads with no JVM configuration at all, therefore relying on the default settings of the HotSpot JVM.  The default settings in OpenJDK are intentionally conservative, designed to work across a wide range of environments and scenarios. However, these defaults often lead to suboptimal resource utilization in cloud-based deployments, where memory and CPU tend to be dedicated for application workloads (use of containers and VMs) but still require intelligent management to maximize efficiency and cost-effectiveness. To address this, we are excited to introduce jaz, a new JVM launcher optimized specifically for …

Do you ever wish you could simply tell your agent to send an email, without the hassle of typing everything — from recipient list to the subject and the body? If so, this guide on building an email-sending agent might be exactly what you’re looking for. Technically, this guide won’t deliver a fully automated agent right out of the box - you’ll still need to add a speech-to-text layer and carefully curate your prompt instructions. By the end of this post, you’ll have an agent capable of interacting with users through natural conversation and generating emails with dynamic subject lines and content. Overview Azure AI Agent Services offers a robust framework for building conversational agents, making it an ideal choice for developers seeking enterprise-grade security and compliance. This ensu…  ( 29 min )

Do you ever wish you could simply tell your agent to send an email, without the hassle of typing everything — from recipient list to the subject and the body? If so, this guide on building an email-sending agent might be exactly what you’re looking for. Technically, this guide won’t deliver a fully automated agent right out of the box - you’ll still need to add a speech-to-text layer and carefully curate your prompt instructions. By the end of this post, you’ll have an agent capable of interacting with users through natural conversation and generating emails with dynamic subject lines and content. Overview Azure AI Agent Services offers a robust framework for building conversational agents, making it an ideal choice for developers seeking enterprise-grade security and compliance. This ensu…

We’re excited to announce that OneLake security capabilities have been extended to support mirrored data through Azure Mirrored Databricks Catalog. This enhancement brings the full suite of OneLake’s enterprise-grade security features to these mirrored assets, empowering organizations to manage access using table, column, or row level security across all engines.  What’s New?  With this update, Azure … Continue reading “Secure Mirrored Azure Databricks Data in Fabric with OneLake security “  ( 6 min )

In our previous blog, we discussed how to debug live issues in the Azure WebJobs Storage Extensions SDK. This approach is particularly effective when issues can be consistently reproduced. However, for intermittent problems, live debugging may not be the most efficient solution. In such cases, integrating custom logging can provide deeper insights and facilitate troubleshooting.   In this blog post, we will provide a step-by-step guide on how to implement custom logging within the Azure WebJobs Storage Extensions SDK. This will help you capture valuable information and better understand the behavior of your applications.   If you encounter any issues while using the Azure WebJobs Extensions SDK, the best way to report them is via GitHub Issues. You can report bugs, request features, or ask…  ( 27 min )

In our previous blog, we discussed how to debug live issues in the Azure WebJobs Storage Extensions SDK. This approach is particularly effective when issues can be consistently reproduced. However, for intermittent problems, live debugging may not be the most efficient solution. In such cases, integrating custom logging can provide deeper insights and facilitate troubleshooting.   In this blog post, we will provide a step-by-step guide on how to implement custom logging within the Azure WebJobs Storage Extensions SDK. This will help you capture valuable information and better understand the behavior of your applications.   If you encounter any issues while using the Azure WebJobs Extensions SDK, the best way to report them is via GitHub Issues. You can report bugs, request features, or ask…

Microsoft just held its annual Microsoft Build event for developers. The live event might be over, but we have highlights and other content that will keep the excitement going. Explore on-demand sessions, learn about recent product announcements, watch deep technical demos, and discover fresh resources for learning cutting-edge developer skills.   Microsoft Build opening keynoteThe world of development—its tools and its possibilities—is rapidly evolving. In the Microsoft Build keynote, Satya Nadella discusses the agentic web, current dev tools, the dev landscape right now, and where it’s headed.   GitHub Copilot: Meet the new coding agentCheck out the exciting new coding agent for GitHub Copilot. Just assign a task or issue to Copilot and it will run in the background, pushing commits to a…  ( 25 min )

Microsoft just held its annual Microsoft Build event for developers. The live event might be over, but we have highlights and other content that will keep the excitement going. Explore on-demand sessions, learn about recent product announcements, watch deep technical demos, and discover fresh resources for learning cutting-edge developer skills.   Microsoft Build opening keynoteThe world of development—its tools and its possibilities—is rapidly evolving. In the Microsoft Build keynote, Satya Nadella discusses the agentic web, current dev tools, the dev landscape right now, and where it’s headed.   GitHub Copilot: Meet the new coding agentCheck out the exciting new coding agent for GitHub Copilot. Just assign a task or issue to Copilot and it will run in the background, pushing commits to a…

Whenever I teach Python workshops, tutorials, or classes, I love to use GitHub Codespaces. Every repository on GitHub can be opened inside a GitHub Codespace, which gives the student a full Python environment and a browser-based VS Code. Students spend less time setting up their environment and more time actually coding - the fun part! In this post, I'll walk through my tips for using Codespaces for teaching Python, particularly for classes about web apps, data science, or generative AI. Getting started You can start a GitHub Codespace from any repository. Navigate to the front page of the repository, then select "Code" > "Codespaces" > "Create codespace on main": By default, the Codespace will build an environment based off a universal Docker image, which includes Python, NodeJS, Java, a…  ( 51 min )

Whenever I teach Python workshops, tutorials, or classes, I love to use GitHub Codespaces. Every repository on GitHub can be opened inside a GitHub Codespace, which gives the student a full Python environment and a browser-based VS Code. Students spend less time setting up their environment and more time actually coding - the fun part! In this post, I'll walk through my tips for using Codespaces for teaching Python, particularly for classes about web apps, data science, or generative AI. Getting started You can start a GitHub Codespace from any repository. Navigate to the front page of the repository, then select "Code" > "Codespaces" > "Create codespace on main": By default, the Codespace will build an environment based off a universal Docker image, which includes Python, NodeJS, Java, a…

The Microsoft Fabric Community Conference is back for its third year—and we’re bringing everything and everybody you’ve loved at past events with us to Atlanta, Georgia. After unforgettable experiences at FabCon in Las Vegas and Stockholm, the Fabric community proved just how powerful it can be when we come together. With more than 13,000 attendees across our last three conferences, it’s clear: the Microsoft Fabric community is here to drive the future of data!    And yes, we’re pleased to announce; it’s happening again! Mark your calendars … Continue reading “Microsoft Fabric Community Conference Comes to Atlanta!”  ( 6 min )

We’re thrilled to announce that we have made Azure Synapse Runtime for Apache Spark 3.5 for our Azure Synapse Spark customers in preview, while they get ready and prepare for migrating to Microsoft Fabric Spark. Apache Spark 3.5 You can now create Azure Synapse Runtime for Apache Spark 3.5. The essential changes include features which come from … Continue reading “Azure Synapse Runtime for Apache Spark 3.5 (Preview)”  ( 5 min )

Following the changes to GitHub Advanced Security on GitHub, we’re launching the standalone security products of GitHub Secret Protection and GitHub Code Security for Azure DevOps today. You can bring the protection of Advanced Security to your enterprise with the flexibility to enable the right level of protection for your repositories. GitHub Secret Protection for […] The post GitHub Secret Protection and GitHub Code Security for Azure DevOps appeared first on Azure DevOps Blog.  ( 23 min )

The pace at which innovative AI models are being developed is outstanding! DeepSeek-R1 is one such model that focuses on complex reasoning tasks, providing a powerful tool for developers to build intelligent applications. The week, we announced its availability on GitHub Models as well as on Azure AI Foundry. In this article, we’ll take a look at how you can deploy and use the DeepSeek-R1 models in your JavaScript applications. TL;DR key takeaways DeepSeek-R1 models focus on complex reasoning tasks, and is not designed for general conversation You can quickly switch your configuration to use Azure AI, GitHub Models, or even local models with Ollama. You can use OpenAI Node SDK or LangChain.js to interact with DeepSeek models. What you'll learn here Deploying DeepSeek-R1 model on Azure. …  ( 32 min )

The pace at which innovative AI models are being developed is outstanding! DeepSeek-R1 is one such model that focuses on complex reasoning tasks, providing a powerful tool for developers to build intelligent applications. The week, we announced its availability on GitHub Models as well as on Azure AI Foundry. In this article, we’ll take a look at how you can deploy and use the DeepSeek-R1 models in your JavaScript applications. TL;DR key takeaways DeepSeek-R1 models focus on complex reasoning tasks, and is not designed for general conversation You can quickly switch your configuration to use Azure AI, GitHub Models, or even local models with Ollama. You can use OpenAI Node SDK or LangChain.js to interact with DeepSeek models. What you'll learn here Deploying DeepSeek-R1 model on Azure. …

Introduction Azure Container Apps provides a serverless platform for running containers at scale, and one of the big benefits is that you can easily scale workloads to zero when they are not getting any traffic. Scaling to zero ensures you only pay when your workloads are actively receiving traffic or performing work. However, for some workloads, scaling to zero might not be possible for a variety of reasons. Some workloads must always be able to respond to requests quickly, and the time it takes to scale from 0 to 1 replicas, while short, is too long. Some applications need to be able to always respond to health checks, and so removing all replicas is not possible. In these scenarios, there may still be time periods where there is no traffic, or the application isn't doing any work. While…  ( 39 min )

Introduction Azure Container Apps provides a serverless platform for running containers at scale, and one of the big benefits is that you can easily scale workloads to zero when they are not getting any traffic. Scaling to zero ensures you only pay when your workloads are actively receiving traffic or performing work. However, for some workloads, scaling to zero might not be possible for a variety of reasons. Some workloads must always be able to respond to requests quickly, and the time it takes to scale from 0 to 1 replicas, while short, is too long. Some applications need to be able to always respond to health checks, and so removing all replicas is not possible. In these scenarios, there may still be time periods where there is no traffic, or the application isn't doing any work. While…

Introduction Ensuring reliable, high-performance serverless applications is central to our work on Azure Functions. With new plans like Flex Consumption expanding the platform’s capabilities, it's critical to continuously validate that our infrastructure can scale—reliably and efficiently—under real-world load. To meet that need, we built PerfBench (Performance Benchmarker), a comprehensive benchmarking system designed to measure, monitor, and maintain our performance baselines—catching regressions before they impact customers. This infrastructure now runs close to 5,000 test executions every month, spanning multiple SKUs, regions, runtimes, and workloads—with Flex Consumption accounting for more than half of the total volume. This scale of testing helps us not only identify regressions ea…  ( 51 min )

Introduction Ensuring reliable, high-performance serverless applications is central to our work on Azure Functions. With new plans like Flex Consumption expanding the platform’s capabilities, it's critical to continuously validate that our infrastructure can scale—reliably and efficiently—under real-world load. To meet that need, we built PerfBench (Performance Benchmarker), a comprehensive benchmarking system designed to measure, monitor, and maintain our performance baselines—catching regressions before they impact customers. This infrastructure now runs close to 5,000 test executions every month, spanning multiple SKUs, regions, runtimes, and workloads—with Flex Consumption accounting for more than half of the total volume. This scale of testing helps us not only identify regressions ea…

Vector embeddings have become very popular over the last few years, but most of us developers are brand new to the concept. In this post, I'll give a high-level overview of embedding models, similarity metrics, vector search, and vector compression approaches. Vector embeddings A vector embedding is a mapping from an input (like a word, list of words, or image) into a list of floating point numbers. That list of numbers represents that input in the multidimensional embedding space of the model. We refer to the length of the list as its dimensions, so a list with 1024 numbers would have 1024 dimensions.   Embedding models Each embedding model has its own dimension length, allowed input types, similarity space, and other characteristics. word2vec For a long time, word2vec was the most well-…  ( 47 min )

Vector embeddings have become very popular over the last few years, but most of us developers are brand new to the concept. In this post, I'll give a high-level overview of embedding models, similarity metrics, vector search, and vector compression approaches. Vector embeddings A vector embedding is a mapping from an input (like a word, list of words, or image) into a list of floating point numbers. That list of numbers represents that input in the multidimensional embedding space of the model. We refer to the length of the list as its dimensions, so a list with 1024 numbers would have 1024 dimensions.   Embedding models Each embedding model has its own dimension length, allowed input types, similarity space, and other characteristics. word2vec For a long time, word2vec was the most well-…

Introduction AI agentic grounding, Retrieval-Augmented Generation (RAG) apps/copilots, and enterprise search are game-changers, but they stay safe only when every response obeys the file-level permissions you already set in your data source. Without built-in help, developers have to hand-code security trimming, unravel nested groups, and tweak the logic whenever someone’s role shifts. Starting with REST API version 2025-05-01-preview, Azure AI Search introduces native Microsoft Entra-based POSIX-style Access Control List (ACL) and Azure Role-Based Access Control (RBAC) support, alongside expanded capabilities in the Azure Data Lake Storage Gen2 (ADLS Gen2) built-in indexer. These enhancements make it easier to enforce document-level security across ingestion and query workflows, whether yo…  ( 40 min )

Introduction AI agentic grounding, Retrieval-Augmented Generation (RAG) apps/copilots, and enterprise search are game-changers, but they stay safe only when every response obeys the file-level permissions you already set in your data source. Without built-in help, developers have to hand-code security trimming, unravel nested groups, and tweak the logic whenever someone’s role shifts. Starting with REST API version 2025-05-01-preview, Azure AI Search introduces native Microsoft Entra-based POSIX-style Access Control List (ACL) and Azure Role-Based Access Control (RBAC) support, alongside expanded capabilities in the Azure Data Lake Storage Gen2 (ADLS Gen2) built-in indexer. These enhancements make it easier to enforce document-level security across ingestion and query workflows, whether yo…

We’re excited to announce that Microsoft Fabric Data Pipelines now support both On-Premises Data Gateway (OPDG) and Virtual Network (VNET) Gateway across a broader set of external activity types. What’s New? You can now securely connect to on-premises and network-isolated resources using OPDG or VNET Gateway for the following Fabric Data Factory pipeline activities allowing for secure … Continue reading “New pipeline Activities Now Support OPDG and VNET”  ( 5 min )

Microsoft Fabric and Azure Databricks are widely used data platforms. This article aims to address the requirement of customers who have large data estates in Azure databricks and want to unlock additional use cases in Microsoft Fabric arising out of different business teams. When integrating the two platforms, a crucial security requirement is to ensure … Continue reading “Integrating Fabric with Databricks using private network”  ( 7 min )

We’re thrilled to introduce Automated Table Statistics in Microsoft Fabric Data Engineering — a major upgrade that helps you get blazing-fast query performance with zero manual effort. Whether you’re running complex joins, large aggregations, or heavy filtering workloads, Fabric’s new automated statistics will help Spark make smarter decisions, saving you time, compute, and money. What … Continue reading “Boost performance effortlessly with Automated Table Statistics in Microsoft Fabric”  ( 6 min )

Step into the future of simplicity with Fabric CLI. The Fabric Command Line Interface (CLI) has arrived in preview, bringing with it a seamless way to create SQL databases for your projects. This guide will take you through the steps to get started, ensuring you can leverage the power of Fabric CLI with ease. Prerequisites Before … Continue reading “How to create a SQL database in Fabric using Fabric CLI”  ( 6 min )

Troubleshooting Azure AI Foundry deployments on Azure App Service  ( 4 min )

In our endeavor to increase coverage of OSS workloads in Azure Migrate, we are announcing discovery and modernization assessment of MySQL databases running on Windows and Linux servers. Customers previously had limited visibility into their MySQL workloads and often received generalized VM lift-and-shift recommendations. With this new capability, customers can now accurately identify their MySQL workloads and assess them for right-sizing into Azure Database for MySQL. MySQL workloads are a cornerstone of the LAMP stack, powering countless web applications with their reliability, performance, and ease of use. As businesses grow, the need for scalable and efficient database solutions becomes paramount. This is where Azure Database for MySQL comes into play. Migrating from on-premises to Azur…  ( 25 min )

In our endeavor to increase coverage of OSS workloads in Azure Migrate, we are announcing discovery and modernization assessment of MySQL databases running on Windows and Linux servers. Customers previously had limited visibility into their MySQL workloads and often received generalized VM lift-and-shift recommendations. With this new capability, customers can now accurately identify their MySQL workloads and assess them for right-sizing into Azure Database for MySQL. MySQL workloads are a cornerstone of the LAMP stack, powering countless web applications with their reliability, performance, and ease of use. As businesses grow, the need for scalable and efficient database solutions becomes paramount. This is where Azure Database for MySQL comes into play. Migrating from on-premises to Azur…

Hi everyone, welcome to the June 2025 update for Jakarta EE and Quarkus on Azure. It covers topics such as DevServices support of Quarkus Azure Extension, and comprehensive guides on implementing Quarkus applications monitoring and Liberty applications monitoring. If you’re interested in providing feedback or collaborating on migrating Java workloads to Azure with the engineering […] The post Jakarta EE and Quarkus on Azure – June 2025 appeared first on Microsoft for Java Developers.  ( 24 min )

We are introducing multiple features in Azure AI Search that make AI agent grounding and RAG data preparation easier. Here is what’s new:  The GenAI prompt skill in public preview accesses Azure AI Foundry chat-completion models to enrich content when it’s being indexed  Logic app integration with AI Search in Azure portal for simple data ingestion  Introduction  Azure AI Search is introducing new features and integrations designed to simplify and accelerate the creation of RAG-ready indexes. The GenAI Prompt Skill leverages generative AI during the indexing process, enabling advanced context expansion, image verbalization, and other transformations to enhance multimodal search relevance. The GenAI Prompt Skill enables sophisticated data transformations during the indexing process and fa…  ( 39 min )

We are introducing multiple features in Azure AI Search that make AI agent grounding and RAG data preparation easier. Here is what’s new:  The GenAI prompt skill in public preview accesses Azure AI Foundry chat-completion models to enrich content when it’s being indexed  Logic app integration with AI Search in Azure portal for simple data ingestion  Introduction  Azure AI Search is introducing new features and integrations designed to simplify and accelerate the creation of RAG-ready indexes. The GenAI Prompt Skill leverages generative AI during the indexing process, enabling advanced context expansion, image verbalization, and other transformations to enhance multimodal search relevance. The GenAI Prompt Skill enables sophisticated data transformations during the indexing process and fa…

We're excited to announce two powerful new enhancements in Azure AI Search: Multi-Vector Field Support and Scoring Profiles Integration with Semantic Ranking. Developed based on your feedback, these features unlock more control and enable additional scenarios in your search experiences   Why these Enhancements Matter As search experiences become increasingly sophisticated, handling complex, multimodal data and maintaining precise relevance is crucial. These new capabilities directly address common pain points: Multi-Vector Field Support helps you manage detailed, multimodal, and segmented content more effectively. Scoring Profiles Integration with Semantic Ranking ensures consistent relevance throughout your search pipeline. Multi-Vector Field Support Previously, vector fields `(Collecti…  ( 27 min )

We're excited to announce two powerful new enhancements in Azure AI Search: Multi-Vector Field Support and Scoring Profiles Integration with Semantic Ranking. Developed based on your feedback, these features unlock more control and enable additional scenarios in your search experiences   Why these Enhancements Matter As search experiences become increasingly sophisticated, handling complex, multimodal data and maintaining precise relevance is crucial. These new capabilities directly address common pain points: Multi-Vector Field Support helps you manage detailed, multimodal, and segmented content more effectively. Scoring Profiles Integration with Semantic Ranking ensures consistent relevance throughout your search pipeline. Multi-Vector Field Support Previously, vector fields `(Collecti…

This is Part 2 of our series on integrating Microsoft.Extensions.AI with Semantic Kernel. In Part 1, we explored the relationship between these technologies and how they complement each other. Now, let’s dive into practical examples showing how to use Microsoft.Extensions.AI abstractions with Semantic Kernel in non-agent scenarios. Getting Started with Microsoft.Extensions.AI and Semantic Kernel Before we […] The post Semantic Kernel and Microsoft.Extensions.AI: Better Together, Part 2 appeared first on Semantic Kernel.  ( 25 min )

The field of AI is rapidly evolving, and the need for more sophisticated, collaborative, and flexible agent-based systems is growing. With this in mind, Semantic Kernel introduces a new multi-agent orchestration framework that enables developers to build, manage, and scale complex agent workflows with ease. This post explores the new orchestration patterns, their capabilities, and […] The post Semantic Kernel: Multi-agent Orchestration appeared first on Semantic Kernel.  ( 26 min )

OneLake allows for security to be defined once and enforced consistently across Microsoft Fabric. One of its standout features is its ability to work seamlessly with shortcuts, offering users the flexibility to access and organize data from different locations while maintaining robust security controls. In this blog post, we will look at how OneLake security … Continue reading “Understanding OneLake Security with Shortcuts”  ( 7 min )

Fabric User Data Functions is a serverless platform that allows you to build and run functions on the Fabric platform. You can use your functions to interact with your data and the rest of your Fabric items via native integrations. After the Fabric User Data Functions preview, we have been working on increasing the number … Continue reading “New regions supported for Fabric User Data Functions”  ( 5 min )

Infrastructure as Code (IaC) tools like Terraform have revolutionized the way developers and organizations deploy and manage infrastructure. With its declarative language and ability to automate provisioning, Terraform reduces human error, ensures consistency, and speeds up deployment across cloud and on-premises environments. In this document, we’ll explore how you can create SQL databases workloads in … Continue reading “Creating SQL database workload in Fabric with Terraform: A Step-by-Step Guide (Preview)”  ( 7 min )

The FinOps toolkit is built on top of the FinOps Framework, providing you a collection of resources to help enterprises facilitate their FinOps Goals. Just announced in the May Updates is an exciting integration with Fabric Real-Time Intelligence. With the latest updates you can now analyze all your cloud spend utilizing Eventhouse. Unlocking the power … Continue reading “Introducing FinOps Toolkit in Fabric”  ( 5 min )

GitHub Copilot has evolved beyond a coding assistant in the IDE into an agentic teammate – providing actionable feedback on pull requests, fixing bugs and implementing new features, creating pull requests and responding to feedback, and much more. These new capabilities will transform every aspect of the software development lifecycle, as we are already seeing […] The post Azure DevOps with GitHub Repositories – Your path to Agentic AI appeared first on Azure DevOps Blog.  ( 26 min )

The GitHub Copilot Global Bootcamp started in February as a fully virtual learning journey — and it was a hit. More than 60,000 developers joined the first edition across multiple languages and regions. Now, we're excited to launch the second edition — bigger and better — featuring both virtual and in-person workshops, hosted by tech communities around the globe. This new edition arrives shortly after the announcements at Microsoft Build 2025, where the GitHub and Visual Studio Code teams revealed exciting news: The GitHub Copilot Chat extension is going open source, reinforcing transparency and collaboration. AI is being deeply integrated into Visual Studio Code, now evolving into an open source AI editor. New APIs and tools are making it easier than ever to build with AI and LLMs. This…  ( 30 min )

The GitHub Copilot Global Bootcamp started in February as a fully virtual learning journey — and it was a hit. More than 60,000 developers joined the first edition across multiple languages and regions. Now, we're excited to launch the second edition — bigger and better — featuring both virtual and in-person workshops, hosted by tech communities around the globe. This new edition arrives shortly after the announcements at Microsoft Build 2025, where the GitHub and Visual Studio Code teams revealed exciting news: The GitHub Copilot Chat extension is going open source, reinforcing transparency and collaboration. AI is being deeply integrated into Visual Studio Code, now evolving into an open source AI editor. New APIs and tools are making it easier than ever to build with AI and LLMs. This…

We’re excited to announce the General Availability (GA) of the Azure Data Factory (Mounting) feature in Microsoft Fabric! This powerful capability allows you to seamlessly connect your existing Azure Data Factory (ADF) pipelines to Fabric workspaces, eliminating the need to manually rebuild or migrate them. What’s New with GA? Once your ADF instance is linked to a Fabric workspace, you … Continue reading “Azure Data Factory Item Mounting (Generally Available)”  ( 5 min )

Summarize, group, and explore data in one step with the new GraphQL aggregations feature We’re excited to launch a powerful new capability in the Fabric API for GraphQL—Aggregations. This feature brings native support for summary-level insights directly into your GraphQL queries, making your data exploration faster, simpler, and more efficient. Why Aggregations? Until now, getting … Continue reading “Introducing Aggregations in Fabric API for GraphQL: Query Smarter, Not Harder”  ( 6 min )

With SQL database in Fabric, the source control integration in Fabric enables you to keep your active work synced to git while following a branching strategy that best matches your team’s environments and deployment requirements. With the complexity of enterprise deployment scenarios, code-first deployment is also available for Fabric objects through tools like the Fabric-CICD … Continue reading “Updates to database development tools for SQL database in Fabric”  ( 6 min )

Co-Author: Maraki Ketema, Principal Product Manager Unlocking Data Value at Scale: Mirroring in Microsoft Fabric  In the modern data era, speed, scale, and simplicity are no longer luxuries—they’re expectations. Organizations want to harness the power of their operational data in real time, without the overhead of complex ETL pipelines or latency-filled data movement. This is … Continue reading “Mirroring in Microsoft Fabric explained: benefits, use cases, and pricing demystified”  ( 9 min )

Turbo charge queries over Delta Lake and Iceberg tables in OneLake Eventhouse accelerated OneLake Table shortcuts aka. Query Acceleration is now Generally Available! OneLake shortcuts are references from an Eventhouse that point to internal Fabric or external sources. Previously, queries run over OneLake shortcuts were less performant than on data that is ingested directly to … Continue reading “Eventhouse Accelerated OneLake Table Shortcuts – Generally Available”  ( 8 min )

In the era of Artificial Intelligence, organizations generate and accumulate large volumes of information every second. From transactional records to user logs and analytics data warehouses serve as a single source of truth that stores this information for a plethora of purposes. However, as data accumulates over time, not all remain relevant and valuable, leading … Continue reading “Intelligent Data Cleanup: Smart Purging for Smarter Data Warehouses”  ( 6 min )

Creating and managing tables in Eventhouse just got even more flexible. While you can create tables using the Get Data wizard – which builds the structure based on a sample file or streaming data – there are times when you need to start from scratch. This could be during basic training, testing, or experimenting with … Continue reading “Eventhouse No-Code table creation and editing”  ( 6 min )

We are happy to announce our public preview for Granular RBAC in Azure Monitor Log Analytics! What is Granular RBAC in Azure Monitor Logs? Many organizations emphasize the need to segregate and control access to data in a fine-grained manner, while maintaining a centralized and consolidated logging platform.  On top of the existing capabilities of workspace and table level access provided over Azure RBAC, you can now maintain all your data in a single Log Analytics workspace and provide least privilege access at any level. This means you can control which users can access which tables and rows, based on your business or security needs and defined criteria, and completely separate data and control plane access, using Azure Attribute-based access control (ABAC) as part of your Azure RBAC role assignment. Granular RBAC in Azure Monitor Logs allows you to filter the data that each user can view or query, based on the conditions that you specify. Common examples are characteristics such as organizational roles and units, geographical locations, or data sensitivity levels.  How to set granular data access in Azure Monitor Logs To set up granular access: Create or edit an Azure role assignment. Under “Conditions”, select “Add condition”. In “Add action”, choose the new DataAction: “Read workspace data”. Under “Build expression”, click “Add expression” to define your access rules. You can use any combination of the “Table Name” and “Column Value” attributes to scope access, leveraging a wide range of supported operators to match your criteria. Once applied, users will only be able to access the data that matches the conditions you've configured.  Get started with Granular RBAC in Azure Monitor Logs Learn more about Granular RBAC and how to set it up in Azure Monitor Logs We hope you enjoy this new addition to Azure Monitor Log Analytics.  ( 21 min )

We are happy to announce our public preview for Granular RBAC in Azure Monitor Log Analytics! What is Granular RBAC in Azure Monitor Logs? Many organizations emphasize the need to segregate and control access to data in a fine-grained manner, while maintaining a centralized and consolidated logging platform.  On top of the existing capabilities of workspace and table level access provided over Azure RBAC, you can now maintain all your data in a single Log Analytics workspace and provide least privilege access at any level. This means you can control which users can access which tables and rows, based on your business or security needs and defined criteria, and completely separate data and control plane access, using Azure Attribute-based access control (ABAC) as part of your Azure RBAC role assignment. Granular RBAC in Azure Monitor Logs allows you to filter the data that each user can view or query, based on the conditions that you specify. Common examples are characteristics such as organizational roles and units, geographical locations, or data sensitivity levels.  How to set granular data access in Azure Monitor Logs To set up granular access: Create or edit an Azure role assignment. Under “Conditions”, select “Add condition”. In “Add action”, choose the new DataAction: “Read workspace data”. Under “Build expression”, click “Add expression” to define your access rules. You can use any combination of the “Table Name” and “Column Value” attributes to scope access, leveraging a wide range of supported operators to match your criteria. Once applied, users will only be able to access the data that matches the conditions you've configured.  Get started with Granular RBAC in Azure Monitor Logs Learn more about Granular RBAC and how to set it up in Azure Monitor Logs We hope you enjoy this new addition to Azure Monitor Log Analytics.

Voice technology is transforming how we interact with machines, making conversations with AI feel more natural than ever before. With the public beta release of the Voice Live API developers now have the tools to create low-latency, multimodal voice experiences in their apps, opening up endless possibilities for innovation. Gone are the days when building a voice bot required stitching together multiple models for transcription, inference, and text-to-speech conversion. With the Realtime API, developers can now streamline the entire process with a single API call, enabling fluid, natural speech-to-speech conversations. This is a game-changer for industries like customer support, education, and real-time language translation, where fast, seamless interactions are crucial. In this blog, we’…  ( 64 min )

Voice technology is transforming how we interact with machines, making conversations with AI feel more natural than ever before. With the public beta release of the Voice Live API developers now have the tools to create low-latency, multimodal voice experiences in their apps, opening up endless possibilities for innovation. Gone are the days when building a voice bot required stitching together multiple models for transcription, inference, and text-to-speech conversion. With the Realtime API, developers can now streamline the entire process with a single API call, enabling fluid, natural speech-to-speech conversations. This is a game-changer for industries like customer support, education, and real-time language translation, where fast, seamless interactions are crucial. In this blog, we’…

Imagine the web as a vast network of pages you click through—until HTML transformed them into living documents you could link and style with simple tags. Announced at Build 2025 during CEO Satya Nadella’s keynote with CTO Kevin Scott, NLWeb is the next leap for conversation—the “HTML for chat”—turning every site into a natural-language endpoint both people and AI agents can query directly. Watch Kevin’s announcement below, or keep reading to learn about our early adopters. NLWeb Pioneer Highlights Tripadvisor: Conversational travel planning, from “Where should I go this fall with kids?” to full itineraries in one go. Read the spotlight. Qdrant: Lightning-fast, intent-aware search via its vector database. Read the spotlight. O’Reilly Media: Queryable technical library, like chatting with a resident expert. Read the spotlight. Eventbrite: Discover events by intent, not keywords. Read the spotlight. Inception Labs: Sub-second conversational queries using diffusion LLMs. Read the spotlight. Delish (Hearst): Instant recipe matches—“quick vegan dinner with mushrooms and pasta”—tailored to your pantry. Read the spotlight. NLWeb instances also double as Model Context Protocol (MCP) servers—exposing your content to AI assistants and enabling peer-to-peer Agent-to-Agent (A2A) workflows across sites. More PioneersChicago Public Media, Common Sense Media, DDM (Allrecipes & Serious Eats), Milvus, Shopify and Snowflake. Together, these collaborators are laying the protocol-based foundation for an open, agentic web—where sites don’t just display content, they converse. Ready to Build?Get started with the new NLWeb GitHub repository.  ( 22 min )

Imagine the web as a vast network of pages you click through—until HTML transformed them into living documents you could link and style with simple tags. Announced at Build 2025 during CEO Satya Nadella’s keynote with CTO Kevin Scott, NLWeb is the next leap for conversation—the “HTML for chat”—turning every site into a natural-language endpoint both people and AI agents can query directly. Watch Kevin’s announcement below, or keep reading to learn about our early adopters. NLWeb Pioneer Highlights Tripadvisor: Conversational travel planning, from “Where should I go this fall with kids?” to full itineraries in one go. Read the spotlight. Qdrant: Lightning-fast, intent-aware search via its vector database. Read the spotlight. O’Reilly Media: Queryable technical library, like chatting with a resident expert. Read the spotlight. Eventbrite: Discover events by intent, not keywords. Read the spotlight. Inception Labs: Sub-second conversational queries using diffusion LLMs. Read the spotlight. Delish (Hearst): Instant recipe matches—“quick vegan dinner with mushrooms and pasta”—tailored to your pantry. Read the spotlight. NLWeb instances also double as Model Context Protocol (MCP) servers—exposing your content to AI assistants and enabling peer-to-peer Agent-to-Agent (A2A) workflows across sites. More PioneersChicago Public Media, Common Sense Media, DDM (Allrecipes & Serious Eats), Milvus, Shopify and Snowflake. Together, these collaborators are laying the protocol-based foundation for an open, agentic web—where sites don’t just display content, they converse. Ready to Build?Get started with the new NLWeb GitHub repository.

Microsoft Build 2025 delivered a powerful vision for the future of data and AI, with Microsoft Fabric and Power BI at the heart of the story. From AI-powered productivity with Copilot to deep integration with Cosmos DB, this year’s announcements reinforced Microsoft’s commitment to unifying the data experience across roles, tools, and industries. Fabric: The … Continue reading “That’s a wrap for Build 2025!”  ( 7 min )

This post is co-authored with Shlomi Elkayam and Henry Hernandez  from Amdocs CCoE. In this blogpost you will learn how Amdocs CCoE team improved their SLA by providing technical support for IT and cloud infrastructure questions and queries. They used Azure AI Agent Service to build an intelligent email agent that helps Amdocs employees with their technical issues. This post will describe the development phases, solution details and the roadmap ahead. About Amdocs CCoE Amdocs is a multinational telecommunications technology company. The company specializes in software and services for communications, media and financial services providers and digital enterprises. CCoE team is responsible for automation, infrastructure and design of all our Azure solutions, either for internal use cases or …  ( 37 min )

This post is co-authored with Shlomi Elkayam and Henry Hernandez  from Amdocs CCoE. In this blogpost you will learn how Amdocs CCoE team improved their SLA by providing technical support for IT and cloud infrastructure questions and queries. They used Azure AI Agent Service to build an intelligent email agent that helps Amdocs employees with their technical issues. This post will describe the development phases, solution details and the roadmap ahead. About Amdocs CCoE Amdocs is a multinational telecommunications technology company. The company specializes in software and services for communications, media and financial services providers and digital enterprises. CCoE team is responsible for automation, infrastructure and design of all our Azure solutions, either for internal use cases or …

As part of the expanding capabilities for agents across Microsoft 365, Office Platform announces add-in actions for Copilot agents are available in preview. This blog is an overview of all the new capabilities across the platform: APIs, developer tools, and add-in distribution options—making it simpler to build new or iterate on JavaScript add-ins. The post Office Add-ins announces Copilot agents with add-in actions and more at Build 2025 appeared first on Microsoft 365 Developer Blog.  ( 31 min )

We’re excited to introduce the Agent Store — a centralized, curated marketplace that features agents built by Microsoft, trusted partners, and customers. The Agent Store offers a new, immersive experience within Microsoft 365 Copilot that enables users to browse, install, and try agents tailored to their needs. The post Introducing the Agent Store: Build, publish, and discover agents in Microsoft 365 Copilot appeared first on Microsoft 365 Developer Blog.  ( 24 min )

With AI-powered capabilities in Data Wrangler, you can now do even more to accelerate exploratory analysis and data preparation in Fabric.  ( 6 min )

The introduction of file-based destinations for Dataflows Gen2 marks a significant step forward in enhancing collaboration, historical tracking, and sharing data for business users. This development begins with SharePoint file destinations in CSV format, offering a streamlined way to share with users on the SharePoint platform. Overview of file-based destinations File-based destinations allow data to … Continue reading “SharePoint files destination the first file-based destination for Dataflows Gen2”  ( 6 min )

Understanding complex data pipelines created by others can often be a challenging task, especially when users must review each activity individually to understand its configurations, settings, and functions. Additionally, manul updates to general settings, such as timeout and retry parameters, across multiple activities can be time-consuming and tedious. Copilot for Data pipeline introduces advanced capabilities … Continue reading “AI-powered development with Copilot for Data pipeline – Boost your productivity in understanding and updating pipeline”  ( 6 min )

During FabCon Las Vegas, we introduced the Fabric CLI — a developer-first command-line tool that brings a familiar, file-system-like experience to working with Microsoft Fabric. Since then, thousands of developers have jumped in: exploring, scripting, and embedding the CLI into local workflows. But for many enterprise teams, one question kept coming up: “When will it … Continue reading “Fabric CLI: explore and automate Microsoft Fabric from your terminal (Generally Available)”  ( 8 min )

As the capabilities of our Fabric Platform Git Integration continue to evolve, we’re happy to share some updates about Fabric’s latest CI/CD capabilities. These advancements are designed to enhance the developer experience and simplify the integration of DevOps practices into everyday workflows. With these innovations, Fabric continues to empower teams to build, test, and deploy … Continue reading “What’s new with Fabric CI/CD – May 2025 “  ( 6 min )

OpenSSL is a powerful tool for working with SSL/TLS, but it can also be used to send custom HTTP requests, which is very useful for debugging and learning how HTTP(S) works at a low level.   1. How It Works Normally, tools like curl or your browser handle everything behind the scenes: they perform the TLS handshake and build the HTTP request. With OpenSSL, you manually create the HTTP request, and OpenSSL only handles the encrypted connection. TLS Handshake: OpenSSL’s s_client command establishes a secure (TLS) connection with the server. Send Request: You pipe (or redirect) a raw HTTP request into s_client. The server processes it and sends a response over the secure channel.   2. Sending a GET Request Here’s how you send a simple GET request to www.bing.com: ( printf "GET / HTTP/1.1\…  ( 23 min )

OpenSSL is a powerful tool for working with SSL/TLS, but it can also be used to send custom HTTP requests, which is very useful for debugging and learning how HTTP(S) works at a low level.   1. How It Works Normally, tools like curl or your browser handle everything behind the scenes: they perform the TLS handshake and build the HTTP request. With OpenSSL, you manually create the HTTP request, and OpenSSL only handles the encrypted connection. TLS Handshake: OpenSSL’s s_client command establishes a secure (TLS) connection with the server. Send Request: You pipe (or redirect) a raw HTTP request into s_client. The server processes it and sends a response over the secure channel.   2. Sending a GET Request Here’s how you send a simple GET request to www.bing.com: ( printf "GET / HTTP/1.1\…

We’re excited to announce the preview of AI powered Actionable Insights in Azure Load Testing—a new capability that helps teams quickly identify performance issues and understand test results through AI-driven analysis.  Performance testing is an essential part of ensuring application reliability and responsiveness, but interpreting the results can often be challenging. It typically involves manually correlating client-side load test telemetry with backend service metrics, which can be both time-consuming and error-prone. Actionable Insights simplifies this process by automatically analyzing test data, surfacing key issues, and offering clear, actionable recommendations—so teams can focus on fixing what matters, not sifting through raw data.  AI-powered diagnostics  Actionable Insights use…  ( 23 min )

We’re excited to announce the preview of AI powered Actionable Insights in Azure Load Testing—a new capability that helps teams quickly identify performance issues and understand test results through AI-driven analysis.  Performance testing is an essential part of ensuring application reliability and responsiveness, but interpreting the results can often be challenging. It typically involves manually correlating client-side load test telemetry with backend service metrics, which can be both time-consuming and error-prone. Actionable Insights simplifies this process by automatically analyzing test data, surfacing key issues, and offering clear, actionable recommendations—so teams can focus on fixing what matters, not sifting through raw data.  AI-powered diagnostics  Actionable Insights use…

Background: We often encounter scenarios where we need to delete blobs that have been idle in a storage account for an extended period. For a small number of blobs, deletion can be handled easily using the Azure Portal, Storage Explorer, or inline scripts such as PowerShell or Azure CLI. However, in most cases, we deal with a large volume of blobs, making manual deletion impractical. In such situations, it's essential to leverage automation tools to streamline the deletion process. One effective option is using Automation Tasks, which can help schedule and manage blob deletions efficiently. Note: Behind the scenes, an automation task is actually a logic app resource that runs a workflow. So, the Consumption pricing model of logic-app applies to automation tasks.    Scenario’s where “Automa…  ( 30 min )

Background: We often encounter scenarios where we need to delete blobs that have been idle in a storage account for an extended period. For a small number of blobs, deletion can be handled easily using the Azure Portal, Storage Explorer, or inline scripts such as PowerShell or Azure CLI. However, in most cases, we deal with a large volume of blobs, making manual deletion impractical. In such situations, it's essential to leverage automation tools to streamline the deletion process. One effective option is using Automation Tasks, which can help schedule and manage blob deletions efficiently. Note: Behind the scenes, an automation task is actually a logic app resource that runs a workflow. So, the Consumption pricing model of logic-app applies to automation tasks.    Scenario’s where “Automa…

Fine-tuning adds new skills to foundational models, simulating experience in the tasks you teach the model to do. This complements Retrieval Augmented Generation, which in real-time uses search to find related information, then add that to your prompts for context. Fine-tuning helps ensure that responses meet your quality expectations for specific repeatable tasks, without needing to be prompting expert. It’s great for drafting complex legal agreements, writing technical documentation, authoring medical papers, and more — using detailed, often lengthy precedent files along with what you teach the model. Using Copilot Studio, anyone can create and deploy these fine-tuned models to use with agents without data science or coding expertise. There, you can teach models using data labeling, grou…  ( 48 min )

Fine-tuning adds new skills to foundational models, simulating experience in the tasks you teach the model to do. This complements Retrieval Augmented Generation, which in real-time uses search to find related information, then add that to your prompts for context. Fine-tuning helps ensure that responses meet your quality expectations for specific repeatable tasks, without needing to be prompting expert. It’s great for drafting complex legal agreements, writing technical documentation, authoring medical papers, and more — using detailed, often lengthy precedent files along with what you teach the model. Using Copilot Studio, anyone can create and deploy these fine-tuned models to use with agents without data science or coding expertise. There, you can teach models using data labeling, grou…

This is the start of a series highlighting the integration between Microsoft Semantic Kernel and Microsoft.Extensions.AI. Future parts will provide detailed examples of using Semantic Kernel with Microsoft.Extensions.AI abstractions.  The most common questions are:  “Do Microsoft’s AI extensions replace Semantic Kernel?”  “When should I use Microsoft’s AI extensions instead of Semantic Kernel?”  This blog post […] The post Semantic Kernel and Microsoft.Extensions.AI: Better Together, Part 1 appeared first on Semantic Kernel.  ( 27 min )

As Semantic Kernel shifts its foundational abstractions to Microsoft.Extensions.AI, we are obsoleting and moving away from our experimental embeddings interfaces to the new standardized abstractions that provide a more consistent and powerful way to work with AI services across the .NET ecosystem. The Evolution of Embedding Generation in Semantic Kernel Semantic Kernel has always aimed […] The post Transitioning to new Extensions AI IEmbeddingGenerator interface appeared first on Semantic Kernel.  ( 23 min )

We’re excited to announce the release of Microsoft.Extensions.VectorData.Abstractions, a foundational library providing exchange types and abstractions for vector stores when working with vector data in AI-powered applications. This release is the result of a close collaboration between the Semantic Kernel and .NET teams, combining expertise in AI and developer tooling to deliver a robust, extensible […] The post Vector Data Extensions are now Generally Available (GA) appeared first on Semantic Kernel.  ( 25 min )

1 Access management Azure Kubernetes Service (AKS) supports Microsoft Entra ID integration, which allows you to control access to your cluster resources using Azure role-based access control (RBAC). In this tutorial, you will learn how to integrate AKS with Microsoft Entra ID and assign different roles and permissions to three types of users: An admin user, who will have full access to the AKS cluster and its resources. A backend ops team, who will be responsible for managing the backend application deployed in the AKS cluster. They will only have access to the backend namespace and the resources within it. A frontend ops team, who will be responsible for managing the frontend application deployed in the AKS cluster. They will only have access to the frontend namespace and the resources wi…  ( 48 min )

1 Access management Azure Kubernetes Service (AKS) supports Microsoft Entra ID integration, which allows you to control access to your cluster resources using Azure role-based access control (RBAC). In this tutorial, you will learn how to integrate AKS with Microsoft Entra ID and assign different roles and permissions to three types of users: An admin user, who will have full access to the AKS cluster and its resources. A backend ops team, who will be responsible for managing the backend application deployed in the AKS cluster. They will only have access to the backend namespace and the resources within it. A frontend ops team, who will be responsible for managing the frontend application deployed in the AKS cluster. They will only have access to the frontend namespace and the resources wi…

AI Developer Hackathon Windows ML  Hosted by Qualcomm on SnapDragonX We’re excited to announce our support and participation for the upcoming global series of Edge AI hackathons, hosted by Qualcomm Technologies. The first is on June 14-15 in Bangalore.  We see a world of hybrid AI, developing rapidly as new generation of intelligent applications get built for diverse scenarios. These range from mobile, desktop, spatial computing and extending all the way to industrial and automotive. Mission critical workloads oscillate between decision-making in the moment, on device, to fine tuning models on the cloud. We believe we are in the early stages of development of agentic applications that efficiently run on the edge for scenarios needing local deployment and on-device inferencing.  Microsoft W…  ( 27 min )

AI Developer Hackathon Windows ML  Hosted by Qualcomm on SnapDragonX We’re excited to announce our support and participation for the upcoming global series of Edge AI hackathons, hosted by Qualcomm Technologies. The first is on June 14-15 in Bangalore.  We see a world of hybrid AI, developing rapidly as new generation of intelligent applications get built for diverse scenarios. These range from mobile, desktop, spatial computing and extending all the way to industrial and automotive. Mission critical workloads oscillate between decision-making in the moment, on device, to fine tuning models on the cloud. We believe we are in the early stages of development of agentic applications that efficiently run on the edge for scenarios needing local deployment and on-device inferencing.  Microsoft W…

GitHub Copilot has changed how developers write code, but when combined with an MCP (Model Copilot Protocol) server, it also connects your services. With it, Copilot can understand your database schema and generate relevant code for your API, data models, or business logic. In this guide, you'll learn how to use the Neon Serverless Postgres MCP server with GitHub Copilot in VS Code to build a sample REST API quickly. We'll walk through how to create an Azure Function that fetches data from a Neon database, all with minimal setup and no manual query writing. From Code Generation to Database Management with GitHub Copilot AI agents are no longer just helping write code—they’re creating and managing databases. When a chatbot logs a customer conversation, or a new region spins up in the Azure …  ( 29 min )

GitHub Copilot has changed how developers write code, but when combined with an MCP (Model Copilot Protocol) server, it also connects your services. With it, Copilot can understand your database schema and generate relevant code for your API, data models, or business logic. In this guide, you'll learn how to use the Neon Serverless Postgres MCP server with GitHub Copilot in VS Code to build a sample REST API quickly. We'll walk through how to create an Azure Function that fetches data from a Neon database, all with minimal setup and no manual query writing. From Code Generation to Database Management with GitHub Copilot AI agents are no longer just helping write code—they’re creating and managing databases. When a chatbot logs a customer conversation, or a new region spins up in the Azure …

Are you unable to delete AI Search with the following error? LockedSPLResourceFound :Unable to verify management locks on Resource '$Resource_Path '. If you still want to delete the search service, manually delete the SPL resource first and try again.  If you are, this is the right place for you to find a quick resolution! Keep on reading through. [Solution – Delete the Shared Private Link] The error message will appear if you still have Shared Private Link configured in AI Search. AI Search will not let you delete the resource unless you delete the Shared Private Link first. You must delete the Shared Private Link in the Portal manually.Move to Settings > Networking > Shared private access tab.     Once the Shared Private Links are all deleted,  please try again to delete the AI Search. Also please give at least 15 minutes for the Shared Private Links to be deleted completely as it may take longer.   [Extra - I tried to delete Shared Private Links but it’s been pending for a long while] There are occasions where you will see your Shared Private Links are in a state of being deleted for a long time as below. (For example 3 hours plus or more) In this case, please open a case to our Support team mentioning you are having an issue with deleting AI Search due to Shared Private Link not being deleted properly. Our team will take care of the issue from that point on!  ( 20 min )

Are you unable to delete AI Search with the following error? LockedSPLResourceFound :Unable to verify management locks on Resource '$Resource_Path '. If you still want to delete the search service, manually delete the SPL resource first and try again.  If you are, this is the right place for you to find a quick resolution! Keep on reading through. [Solution – Delete the Shared Private Link] The error message will appear if you still have Shared Private Link configured in AI Search. AI Search will not let you delete the resource unless you delete the Shared Private Link first. You must delete the Shared Private Link in the Portal manually.Move to Settings > Networking > Shared private access tab.     Once the Shared Private Links are all deleted,  please try again to delete the AI Search. Also please give at least 15 minutes for the Shared Private Links to be deleted completely as it may take longer.   [Extra - I tried to delete Shared Private Links but it’s been pending for a long while] There are occasions where you will see your Shared Private Links are in a state of being deleted for a long time as below. (For example 3 hours plus or more) In this case, please open a case to our Support team mentioning you are having an issue with deleting AI Search due to Shared Private Link not being deleted properly. Our team will take care of the issue from that point on!

Here is the May 2025 release of the on-premises data gateway (version 3000.270).  ( 6 min )

“One Ring to rule them all, One Ring to find them, One Ring to bring them all, and in the darkness bind them.” – J.R.R. Tolkien, The Lord of the Rings In the world of code scanning and dependency scanning, your pipeline is the One Ring—a single definition that can orchestrate scans across multiple repositories. […] The post One Pipeline to Rule Them All: Ensuring CodeQL Scanning Results and Dependency Scanning Results Go to the Intended Repository appeared first on Azure DevOps Blog.  ( 26 min )

Read how the Microsoft 365 Agents Toolkit, an evolution of Microsoft Teams Toolkit, is designed to help developers build agents and apps for Microsoft 365 Copilot, Microsoft Teams, and Microsoft 365. The post Introducing the Microsoft 365 Agents Toolkit appeared first on Microsoft 365 Developer Blog.  ( 24 min )

Imagine running advanced AI applications—like intelligent copilots and Retrieval-Augmented Generation (RAG)—directly on Android devices, completely offline. With the rapid evolution of Neural Processing Units (NPUs), this is no longer a future vision—it’s happening now. Optimized AI at the Edge: Phi-4-mini on MediaTek Thanks to MediaTek’s conversion and quantization tools, Microsoft’s Phi-4-mini and Phi-4-mini-reasoning models are now optimized for MediaTek NPUs. This collaboration empowers developers to build fast, responsive, and privacy-preserving AI experiences on Android—without needing cloud connectivity. MediaTek’s flagship Dimensity 9400 and 9400+ platform with Dimensity GenAI Toolkit 2.0 delivers excellent performance with the Phi-4 mini (3.8B) model where ​prefill speed is >800 t…  ( 29 min )

Imagine running advanced AI applications—like intelligent copilots and Retrieval-Augmented Generation (RAG)—directly on Android devices, completely offline. With the rapid evolution of Neural Processing Units (NPUs), this is no longer a future vision—it’s happening now. Optimized AI at the Edge: Phi-4-mini on MediaTek Thanks to MediaTek’s conversion and quantization tools, Microsoft’s Phi-4-mini and Phi-4-mini-reasoning models are now optimized for MediaTek NPUs. This collaboration empowers developers to build fast, responsive, and privacy-preserving AI experiences on Android—without needing cloud connectivity. MediaTek’s flagship Dimensity 9400 and 9400+ platform with Dimensity GenAI Toolkit 2.0 delivers excellent performance with the Phi-4 mini (3.8B) model where ​prefill speed is >800 t…

We are seeing customers such as Indiana Pacers and Coca-Cola transform customer experiences using Azure AI Speech to power customer interactions. And in the new era of agentic AI, voice is increasingly becoming an important modality to interact with AI Agents in a natural way. Today, we are excited to announce a number to new capabilities in Azure AI Speech that will further propel our customers in the voice-enabled agentic AI era as AI Agents are being rapidly adopted by a wide range of enterprise customers across a wide variety of industries. The updates we are announcing today include the new Voice Live API (Public Preview) which can be used to help simplify creating voice agents that provide fluent and natural speech to speech conversational experiences. To provide a robust conversatio…  ( 47 min )

We are seeing customers such as Indiana Pacers and Coca-Cola transform customer experiences using Azure AI Speech to power customer interactions. And in the new era of agentic AI, voice is increasingly becoming an important modality to interact with AI Agents in a natural way. Today, we are excited to announce a number to new capabilities in Azure AI Speech that will further propel our customers in the voice-enabled agentic AI era as AI Agents are being rapidly adopted by a wide range of enterprise customers across a wide variety of industries. The updates we are announcing today include the new Voice Live API (Public Preview) which can be used to help simplify creating voice agents that provide fluent and natural speech to speech conversational experiences. To provide a robust conversatio…

“We’re excited to be part of Microsoft Build and to demonstrate what’s possible when AI meets every day tech. Built on deep technical integration and shared innovation goals, we’re able to deliver smarter, more intuitive, and responsive audio products for users around the world.”— Dongping Zhao, President of Anker Innovations   Imagine talking to anyone, no matter the language. soundcore, Anker Innovations' audio brand, has incorporated Microsoft Azure AI Speech services into its new devices to eliminate language barriers. These wireless earbuds now offer real-time speech translation and voice interactions, showcasing how cloud-based AI speech technologies can create immersive, multilingual experiences on consumer devices.    Anker’s Mission and Challenges Anker Innovations is a global sma…  ( 28 min )

“We’re excited to be part of Microsoft Build and to demonstrate what’s possible when AI meets every day tech. Built on deep technical integration and shared innovation goals, we’re able to deliver smarter, more intuitive, and responsive audio products for users around the world.”— Dongping Zhao, President of Anker Innovations   Imagine talking to anyone, no matter the language. soundcore, Anker Innovations' audio brand, has incorporated Microsoft Azure AI Speech services into its new devices to eliminate language barriers. These wireless earbuds now offer real-time speech translation and voice interactions, showcasing how cloud-based AI speech technologies can create immersive, multilingual experiences on consumer devices.    Anker’s Mission and Challenges Anker Innovations is a global sma…

As a Product Marketing Director at Azure, I’ve had a front-row seat to the evolution of generative AI—from early text-based bots to today’s intelligent systems that reason across images, documents, and real-world context. But some of the most exciting shifts aren’t just about models—they’re about the web itself. Enter NLWeb, Microsoft’s newly announced open initiative to make websites conversational and AI-native. Imagine asking a recipe site like Delish, “What’s a quick, vegan dinner I can make with mushrooms and pasta?”—and getting a smart, tailored response that pulls directly from your structured content, without relying on rigid keyword search.  Built on familiar tools like Schema.org and vector search, NLWeb is designed to be easy for developers and impactful for users. It’s led by R.V. Guha—the mind behind Schema.org, RSS, and RDF—who recently joined Microsoft to help reimagine the open web for the AI era. The new GitHub repo is now live for developers to explore and build upon. Here’s how NLWeb pioneer Delish is thinking about this shift: Q1: What inspired your team to try NLWeb?  We saw an opportunity to improve discovery for consumers by delivering more relevant results through faceted, natural language queries. Q2: How did the setup process go? Any surprises? Collaborating with the Microsoft engineers was valuable during the test planning, and the initial prototype results were promising as we explored the potential. Q3: What query or interaction made NLWeb click for you?  Users can ask naturally phrased questions—based on cultural moments or food types—and get accurate (and delicious) results. Q4: How are you blending NLWeb with your current experience?  We will actively be testing NLWeb embedded as part of the discovery experience on the Delish site. Q5: If NLWeb reaches its full potential, what could it unlock for your users or the web?  We’re excited about the potential to better serve our customers, drive deeper engagement, increase time spent and grow higher LTV audiences.  ( 21 min )

As a Product Marketing Director at Azure, I’ve had a front-row seat to the evolution of generative AI—from early text-based bots to today’s intelligent systems that reason across images, documents, and real-world context. But some of the most exciting shifts aren’t just about models—they’re about the web itself. Enter NLWeb, Microsoft’s newly announced open initiative to make websites conversational and AI-native. Imagine asking a recipe site like Delish, “What’s a quick, vegan dinner I can make with mushrooms and pasta?”—and getting a smart, tailored response that pulls directly from your structured content, without relying on rigid keyword search.  Built on familiar tools like Schema.org and vector search, NLWeb is designed to be easy for developers and impactful for users. It’s led by R.V. Guha—the mind behind Schema.org, RSS, and RDF—who recently joined Microsoft to help reimagine the open web for the AI era. The new GitHub repo is now live for developers to explore and build upon. Here’s how NLWeb pioneer Delish is thinking about this shift: Q1: What inspired your team to try NLWeb?  We saw an opportunity to improve discovery for consumers by delivering more relevant results through faceted, natural language queries. Q2: How did the setup process go? Any surprises? Collaborating with the Microsoft engineers was valuable during the test planning, and the initial prototype results were promising as we explored the potential. Q3: What query or interaction made NLWeb click for you?  Users can ask naturally phrased questions—based on cultural moments or food types—and get accurate (and delicious) results. Q4: How are you blending NLWeb with your current experience?  We will actively be testing NLWeb embedded as part of the discovery experience on the Delish site. Q5: If NLWeb reaches its full potential, what could it unlock for your users or the web?  We’re excited about the potential to better serve our customers, drive deeper engagement, increase time spent and grow higher LTV audiences.

First introduced in public preview last year, Azure AI Content Understanding enables you to convert unstructured content—documents, audio, video, text, and images—into structured data. The service is designed to support consistent, high-quality output, directed improvements, built-in enrichment, and robust pre-processing to accelerate workflows and reduce cost. A New Chapter in Content Understanding Since our launch we’ve seen customers pushing the boundaries to go beyond simple data extraction with agentic solutions fully automating decisions. This requires more than just extracting fields. For example, a healthcare insurance provider decision to pay a claim requires cross-checking against insurance policies, applicable contracts, patient’s medical history and prescription datapoints. To …  ( 32 min )

First introduced in public preview last year, Azure AI Content Understanding enables you to convert unstructured content—documents, audio, video, text, and images—into structured data. The service is designed to support consistent, high-quality output, directed improvements, built-in enrichment, and robust pre-processing to accelerate workflows and reduce cost. A New Chapter in Content Understanding Since our launch we’ve seen customers pushing the boundaries to go beyond simple data extraction with agentic solutions fully automating decisions. This requires more than just extracting fields. For example, a healthcare insurance provider decision to pay a claim requires cross-checking against insurance policies, applicable contracts, patient’s medical history and prescription datapoints. To …

We just announced NLWeb at Microsoft Build—starting with a GitHub repo to help developers explore it, and a short list of enterprise pioneers testing it out in the real world. Qdrant is one of those early innovators shaping where this goes. Known for their open-source vector database purpose-built for semantic search, Qdrant is helping developers supercharge the intelligence of their search interfaces—without rebuilding their entire stack. By integrating with NLWeb, Qdrant makes it easy to add fast, intent-aware, and context-rich search to websites and apps of any size. Below, the Qdrant team shares how this integration came together, what developers can expect, and why NLWeb might be the unlock that brings semantic search to the mainstream.     Q1: Why Qdrant Sees NLWeb as a Practical St…  ( 27 min )

We just announced NLWeb at Microsoft Build—starting with a GitHub repo to help developers explore it, and a short list of enterprise pioneers testing it out in the real world. Qdrant is one of those early innovators shaping where this goes. Known for their open-source vector database purpose-built for semantic search, Qdrant is helping developers supercharge the intelligence of their search interfaces—without rebuilding their entire stack. By integrating with NLWeb, Qdrant makes it easy to add fast, intent-aware, and context-rich search to websites and apps of any size. Below, the Qdrant team shares how this integration came together, what developers can expect, and why NLWeb might be the unlock that brings semantic search to the mainstream.     Q1: Why Qdrant Sees NLWeb as a Practical St…

We just announced NLWeb at Microsoft Build—starting with a GitHub repo to help developers explore it, and a short list of enterprise pioneers testing it out in the real world. Eventbrite is one of those early innovators shaping where this goes. Known for their global events platform that helps millions discover and attend unique experiences, Eventbrite is now exploring how NLWeb can make that discovery process even more personal, conversational, and precise. Whether you're looking for a date night idea or planning your next creative outing, NLWeb enables natural, expressive search queries that understand intent—not just keywords. Below, Eventbrite shares their journey piloting NLWeb: what made them say yes, how it’s performing in early tests, and what they see on the horizon. Q1: What ins…  ( 27 min )

We just announced NLWeb at Microsoft Build—starting with a GitHub repo to help developers explore it, and a short list of enterprise pioneers testing it out in the real world. Eventbrite is one of those early innovators shaping where this goes. Known for their global events platform that helps millions discover and attend unique experiences, Eventbrite is now exploring how NLWeb can make that discovery process even more personal, conversational, and precise. Whether you're looking for a date night idea or planning your next creative outing, NLWeb enables natural, expressive search queries that understand intent—not just keywords. Below, Eventbrite shares their journey piloting NLWeb: what made them say yes, how it’s performing in early tests, and what they see on the horizon. Q1: What ins…

As a Director of AI Product Marketing at Azure, I have spent the last three years deep in the GenAI ecosystem. From internal chatbots that evolved into multi-model agent orchestrations, I feel like I am witnessing history in my job and life. Every day I wonder what is coming next, to build on top of everything we have developed in just a short amount of time. Now I am excited to share the newest chapter in this incredible run of innovation with the announcement of Microsoft’s launch of NLWeb today, an open project aimed at facilitating the creation of natural language interfaces for websites, enabling users to interact with site content through natural language queries. The initiative strives to empower web publishers by making it easier to develop AI-driven applications that enhance user …  ( 24 min )

As a Director of AI Product Marketing at Azure, I have spent the last three years deep in the GenAI ecosystem. From internal chatbots that evolved into multi-model agent orchestrations, I feel like I am witnessing history in my job and life. Every day I wonder what is coming next, to build on top of everything we have developed in just a short amount of time. Now I am excited to share the newest chapter in this incredible run of innovation with the announcement of Microsoft’s launch of NLWeb today, an open project aimed at facilitating the creation of natural language interfaces for websites, enabling users to interact with site content through natural language queries. The initiative strives to empower web publishers by making it easier to develop AI-driven applications that enhance user …

Over the past three years working in generative AI at Microsoft, I’ve had a front-row seat to some of the most exciting shifts in tech. From chat assistants to multi-modal agents, every layer we’ve built has opened up new questions—and new responsibilities. One of the biggest ones: how do we make the web itself more useful to humans and AI? Enter NLWeb—a new open project from Microsoft that aims to make websites natively conversational. Instead of keyword searches or rigid menus, users (and agents) can query a site’s content in natural language. For developers and web publishers, NLWeb offers a practical way to tap into your existing structured data—like Schema.org markup or product catalogs—and expose that content intelligently, without reinventing your site. I’m especially excited that O…  ( 27 min )

Over the past three years working in generative AI at Microsoft, I’ve had a front-row seat to some of the most exciting shifts in tech. From chat assistants to multi-modal agents, every layer we’ve built has opened up new questions—and new responsibilities. One of the biggest ones: how do we make the web itself more useful to humans and AI? Enter NLWeb—a new open project from Microsoft that aims to make websites natively conversational. Instead of keyword searches or rigid menus, users (and agents) can query a site’s content in natural language. For developers and web publishers, NLWeb offers a practical way to tap into your existing structured data—like Schema.org markup or product catalogs—and expose that content intelligently, without reinventing your site. I’m especially excited that O…

Over the past three years working in generative AI at Microsoft, I’ve had a front-row seat to some of the most exciting shifts in tech. From chat assistants to multi-modal agents, every layer we’ve built has opened up new questions—and new responsibilities. One of the biggest ones: how do we make the web itself more useful to humans and AI? Enter NLWeb—a new open project from Microsoft that aims to make websites natively conversational. Instead of keyword searches or rigid menus, users (and agents) can query a site’s content in natural language. For developers and web publishers, NLWeb offers a practical way to tap into your existing structured data—like Schema.org markup or product catalogs—and expose that content intelligently, without reinventing your site. I’m especially excited that O…

As a Product Marketing Director at Azure, I’ve spent the last few years in the thick of generative AI, from early chatbot experiments to sophisticated agentic systems that now reason across modalities. It’s felt like living inside a tech documentary, with every month adding a new chapter. Today, I’m excited to share what might be one of the most quietly profound shifts yet: NLWeb. Just announced by Microsoft, NLWeb is an open initiative to make the entire web more conversational—enabling users (and AI agents) to interact with websites via natural language instead of dropdowns and keyword searches. For developers and publishers, NLWeb simplifies how you expose your content to AI by using structured data formats like Schema.org, vector indexes, and LLM tools you probably already use. What ma…  ( 29 min )

As a Product Marketing Director at Azure, I’ve spent the last few years in the thick of generative AI, from early chatbot experiments to sophisticated agentic systems that now reason across modalities. It’s felt like living inside a tech documentary, with every month adding a new chapter. Today, I’m excited to share what might be one of the most quietly profound shifts yet: NLWeb. Just announced by Microsoft, NLWeb is an open initiative to make the entire web more conversational—enabling users (and AI agents) to interact with websites via natural language instead of dropdowns and keyword searches. For developers and publishers, NLWeb simplifies how you expose your content to AI by using structured data formats like Schema.org, vector indexes, and LLM tools you probably already use. What ma…

In today’s fast-moving AI landscape, businesses are racing to embed conversational intelligence and automation into every customer touchpoint. However, building a reliable and scalable agent from scratch remains complex and time-consuming. Developers tell us they need a streamlined way to map diverse user intents, craft accurate responses, and support global audiences without wrestling with ad-hoc integrations. At the same time, rising expectations around data privacy and compliance introduce yet another layer of overhead. To meet these challenges, today, we’re excited to announce a suite of powerful new tools and templates designed to help developers build intelligent agents faster than ever with our Azure AI Language service. Working together with Azure AI Agent Service, whether you’re t…  ( 37 min )

We're excited to announce Sora, now available in AI Foundry. Learn more about Sora in Azure OpenAI, and its API and video playground, here. To dive deeper into video playground experience, check out this blog post ​  Why does this matter?  AI is no longer *just* about text. Here’s why: multimodal models enable deeper understanding. Today AI doesn’t just understand words: it understands context, visuals, and motion. From prompt to product, imagine going from a product description to a full marketing campaign.  In many ways, the rise of multimodal AI today is comparable to the inception of photography in the 19th century—introducing a new creative medium that, like photography, didn’t replace painting but expanded the boundaries of artistic expression. Just a little over a month ago, we rele…  ( 29 min )

A New Era of Agent Intelligence We’re thrilled to announce the public preview of Tracing, Evaluation, and Monitoring in Azure AI Foundry Agent Service, features designed to revolutionize how developers build, debug, and optimize AI agents. With detailed traces and customizable evaluators, AgentOps is here to bridge the gap between observability and performance improvement. Whether you’re managing a simple chatbot or a complex multi-agent system, this is the tool you’ve been waiting for. What Makes AgentOps Unique? AgentOps offers an unparalleled suite of functionalities that cater to the challenges AI developers face today. Here are the two cornerstone features: 1. Integrated Tracing Functionality AgentOps provides full execution tracing, offering a detailed, step-by-step breakdown of how…  ( 25 min )

If you’re building an AI agent and need to customize its behavior to specific domains, its interaction tone, or improve its tool selection, you should be fine tuning! Our customers agree, but the challenge faced has been twofold: regional availability and the cost of experimentation. Today, we’re bringing fine tuning of Azure OpenAI models to a dozen new AI Foundry regions with the public previews of Global Training and Developer Tier. With reduced pricing and global availability, AI Foundry makes fine tuning the latest OpenAI models on Azure more accessible and affordable than ever to bring your agents to life. What is Global Training? Global Training expands the reach of model customization with the affordable pricing of our other Global offerings: 🏋️‍♂️ Train the latest OpenAI models …  ( 27 min )

Customers often face challenges deploying AI agents that can handle complex, real-world scenarios due to limited tool access. To empower AI agents with more capabilities, Azure AI Foundry Agent Service now supports a growing set of integrated tools: Grounding with Bing Custom Search (preview), SharePoint (coming soon), Azure Logic Apps (preview), Triggers (preview), and third-party tools, enabling agents to retrieve richer information, take more meaningful actions, and deliver intelligent, goal-driven results. This expansion helps organizations build smarter, more adaptable agents that can better align with dynamic business needs.   Today, we are thrilled to announce Azure AI Foundry Agent Service is now Generally Available including following tools: Azure AI Search, Azure Functions, Code …  ( 43 min )

As organizations increasingly rely on AI to automate complex tasks and scale digital operations, the ability to coordinate multiple agents in a single, cohesive system is becoming critical. Moving beyond single-agent architectures to multi-agent systems enables richer, more dynamic automation – where specialized agents can collaborate, share context, and complete multi-step processes with minimal human intervention. This shift is unlocking the potential for organizations to build full digital workforces that can manage everything from customer support to supply chain automation.   Why Multi-Agents Matter Building a single agent is often straightforward – it’s designed to perform a specific task, like answering common support queries or generating summaries from documents. However, real-wor…  ( 49 min )

Agents are revolutionizing business automation by evolving from simple chatbots into sophisticated, collaborative systems. Fueled by advancements in model reasoning and efficiency, these agents can now handle complex, multi-step processes with speed and accuracy. This marks a shift from isolated tools to dynamic, scalable agent workforces that coordinate tasks, share context, and adapt in real time. This transformation enables businesses to optimize operations and elevate customer experiences through AI-driven workflows. Instead of relying on single-purpose bots, organizations are deploying ecosystems of specialized agents that can interact, reason, and respond to changing conditions with minimal oversight. Yet building and scaling these systems is not without challenges. Developers must i…  ( 39 min )

In today’s fast-moving AI landscape, businesses are racing to embed conversational intelligence and automation into every customer touchpoint. However, building a reliable and scalable agent from scratch remains complex and time-consuming. Developers tell us they need a streamlined way to map diverse user intents, craft accurate responses, and support global audiences without wrestling with ad-hoc integrations. At the same time, rising expectations around data privacy and compliance introduce yet another layer of overhead. To meet these challenges, today, we’re excited to announce a suite of powerful new tools and templates designed to help developers build intelligent agents faster than ever with our Azure AI Language service. Working together with Azure AI Agent Service, whether you’re t…

We're excited to announce Sora, coming soon to AI Foundry. Learn more about Sora in Azure OpenAI, and its API and video playground, here.  Hear how T&Pm, a WPP company, uses Sora in Azure OpenAI to bring their media solutions to the next level   AI is no longer *just* about text. Here’s why: multimodal models enable deeper understanding. Today AI doesn’t just understand words – it understands context, visuals, and motion. From prompt to product, imagine going from a product description to a full marketing campaign.  This shift represents more than just a technical evolution—it’s a creative revolution. In many ways, the rise of multimodal AI today is comparable to the inception of photography in the 19th century—introducing a new creative medium that, like photography, didn’t replace pain…

A New Era of Agent Intelligence We’re thrilled to announce the public preview of Tracing, Evaluation, and Monitoring in Azure AI Foundry Agent Service, features designed to revolutionize how developers build, debug, and optimize AI agents. With detailed traces and customizable evaluators, AgentOps is here to bridge the gap between observability and performance improvement. Whether you’re managing a simple chatbot or a complex multi-agent system, this is the tool you’ve been waiting for. What Makes AgentOps Unique? AgentOps offers an unparalleled suite of functionalities that cater to the challenges AI developers face today. Here are the two cornerstone features: 1. Integrated Tracing Functionality AgentOps provides full execution tracing, offering a detailed, step-by-step breakdown of how…

If you’re building an AI agent and need to customize its behavior to specific domains, its interaction tone, or improve its tool selection, you should be fine tuning! Our customers agree, but the challenge faced has been twofold: regional availability and the cost of experimentation. Today, we’re bringing fine tuning of Azure OpenAI models to a dozen new AI Foundry regions with the public previews of Global Training and Developer Tier. With reduced pricing and global availability, AI Foundry makes fine tuning the latest OpenAI models on Azure more accessible and affordable than ever to bring your agents to life. What is Global Training? Global Training expands the reach of model customization with the affordable pricing of our other Global offerings: 🏋️‍♂️ Train the latest OpenAI models …

Customers often face challenges deploying AI agents that can handle complex, real-world scenarios due to limited tool access. To empower AI agents with more capabilities, Azure AI Foundry Agent Service now supports a growing set of integrated tools: Grounding with Bing Custom Search (preview), SharePoint (coming soon), Azure Logic Apps (preview), Triggers (preview), and third-party tools, enabling agents to retrieve richer information, take more meaningful actions, and deliver intelligent, goal-driven results. This expansion helps organizations build smarter, more adaptable agents that can better align with dynamic business needs.   Today, we are thrilled to announce Azure AI Foundry Agent Service is now Generally Available including following tools: Azure AI Search, Azure Functions, Code …

As organizations increasingly rely on AI to automate complex tasks and scale digital operations, the ability to coordinate multiple agents in a single, cohesive system is becoming critical. Moving beyond single-agent architectures to multi-agent systems enables richer, more dynamic automation – where specialized agents can collaborate, share context, and complete multi-step processes with minimal human intervention. This shift is unlocking the potential for organizations to build full digital workforces that can manage everything from customer support to supply chain automation.   Why Multi-Agents Matter Building a single agent is often straightforward – it’s designed to perform a specific task, like answering common support queries or generating summaries from documents. However, real-wor…

Agents are revolutionizing business automation by evolving from simple chatbots into sophisticated, collaborative systems. Fueled by advancements in model reasoning and efficiency, these agents can now handle complex, multi-step processes with speed and accuracy. This marks a shift from isolated tools to dynamic, scalable agent workforces that coordinate tasks, share context, and adapt in real time. This transformation enables businesses to optimize operations and elevate customer experiences through AI-driven workflows. Instead of relying on single-purpose bots, organizations are deploying ecosystems of specialized agents that can interact, reason, and respond to changing conditions with minimal oversight. Yet building and scaling these systems is not without challenges. Developers must i…

Optimizing JVM Configuration for Azure Deployments Tuning the Java Virtual Machine (JVM) for cloud deployments is notoriously challenging. Over 30% of developers tend to deploy Java workloads with no JVM configuration at all, therefore relying on the default settings of the HotSpot JVM. The default settings in OpenJDK are intentionally conservative, designed to work across […] The post Announcing Azure Command Launcher for Java appeared first on Microsoft for Java Developers.  ( 24 min )

Today, we’re excited to announce that GitHub Copilot Agent Mode and MCP support are now in public preview for both JetBrains and Eclipse! Whether you’re working in IntelliJ IDEA, PyCharm, WebStorm or Eclipse, you can now access Copilot’s intelligent agent features and seamlessly manage your project workflows, all from within your IDE.  In this post, we’ll […] The post Vibe coding with GitHub Copilot: Agent mode and MCP support in JetBrains and Eclipse appeared first on Microsoft for Java Developers.  ( 24 min )

In the era of big data, efficiently parsing and analyzing JSON data is critical for gaining actionable insights. Leveraging Kusto, a powerful query engine developed by Microsoft, enhances the efficiency of handling JSON data, making it simpler and faster to derive meaningful patterns and trends. Perhaps more importantly, Kusto’s ability to easily parse simple or … Continue reading “Efficient JSON loading to Eventhouse in Fabric Real-Time Intelligence”  ( 8 min )

Copy job is designed to simplify your data ingestion experience without compromise from any source to any destination. It supports multiple data delivery styles, including both batch and incremental copy, providing the flexibility to meet diverse needs. We are excited to introduce the preview of native Change Data Capture (CDC) support in Copy job that … Continue reading “Simplifying Data Ingestion with Copy job – Introducing Change Data Capture (CDC) Support (Preview)”  ( 7 min )

Dataflows have great capabilities to help you add new columns. It has an extensive section in the ribbon that helps you create new columns based on the data of your table. For scenarios in which you wish to have full control over how the new column should be created, you can opt for the Custom … Continue reading “New Copilot experience in Dataflow Gen2: Natural language to custom column”  ( 6 min )

The integration of Azure Storage with Fabric Eventhouse for continuous ingestion represents a significant simplification of data ingestion process from Azure Storage to Eventhouse in Fabric Real-Time Intelligence. It automates extraction and loading from Azure storage and facilitates near real-time updates to Eventhouse KQL DB tables. With this feature, it is now easier for organizations … Continue reading “Continuous Ingestion from Azure Storage to Eventhouse (Preview)”  ( 9 min )

Co-author: Joanne Wong We’re excited to announce the upcoming integration of Fabric data agent with Copilot in Power BI, enhancing your ability to extract insights seamlessly. What’s new? A new chat with your data experience is launching soon in Power BI– a full-screen Copilot for users to ask natural language questions and receive accurate, relevant … Continue reading “Extracting deeper insights with Fabric Data Agents in Copilot in Power BI”  ( 7 min )

In this blog post we’ll cover the process of integrating Key Vault in the Java Spring Boot app that runs on App Service Linux.  ( 6 min )

Spring AI 1.0 is now generally available, and it is ready to help Java developers bring the power of AI into their Spring Boot applications. This release is the result of open collaboration and contributions across the Spring and Microsoft Azure engineering teams. Together, they have made it simple for Java developers to integrate LLMs, vector search, memory, and agentic workflows using the patterns they already know. Why This Matters for Java Developers? Spring AI 1.0, built and maintained by the Spring team at Broadcom with active contributions from Microsoft Azure, delivers an intuitive and powerful foundation for building intelligent apps. You can plug AI into existing Spring Boot apps with minimal friction, using starters and conventions familiar to every Spring developer. Whether you…  ( 32 min )

Spring AI 1.0 is now generally available, and it is ready to help Java developers bring the power of AI into their Spring Boot applications. This release is the result of open collaboration and contributions across the Spring and Microsoft Azure engineering teams. Together, they have made it simple for Java developers to integrate LLMs, vector search, memory, and agentic workflows using the patterns they already know. Why This Matters for Java Developers? Spring AI 1.0, built and maintained by the Spring team at Broadcom with active contributions from Microsoft Azure, delivers an intuitive and powerful foundation for building intelligent apps. You can plug AI into existing Spring Boot apps with minimal friction, using starters and conventions familiar to every Spring developer. Whether you…

Today we're excited to announce the public preview of Red Hat OpenShift Virtualization on Azure Red Hat OpenShift (ARO). This new capability brings you:  The best of both worlds - Run your virtual machines and containers on a single platform with unified management  Modernization at your pace - Keep critical VMs running while gradually moving components to containers when you're ready  Maximum value from Azure - Leverage your existing Azure benefits and commitments while optimizing resource usage across all workloads  This collaboration between Microsoft and Red Hat helps you simplify operations, reduce costs, and accelerate your cloud journey - all while preserving your existing VM investments.  Why Organizations Need a Unified VM and Container Platform  Organizations today face sig…  ( 39 min )

Today we're excited to announce the public preview of Red Hat OpenShift Virtualization on Azure Red Hat OpenShift (ARO). This new capability brings you:  The best of both worlds - Run your virtual machines and containers on a single platform with unified management  Modernization at your pace - Keep critical VMs running while gradually moving components to containers when you're ready  Maximum value from Azure - Leverage your existing Azure benefits and commitments while optimizing resource usage across all workloads  This collaboration between Microsoft and Red Hat helps you simplify operations, reduce costs, and accelerate your cloud journey - all while preserving your existing VM investments.  Why Organizations Need a Unified VM and Container Platform  Organizations today face sig…

Overview This blog post shows you how to enable SmallRye Health in Quarkus applications with health probes in Azure Container Apps. The techniques shown in this blog post show you how to effectively monitor the health and status of your Quarkus instances. The application is a "to do list" with a JavaScript front end and a REST endpoint. Azure Database for PostgreSQL Flexible Server provides the persistence layer for the app. The app uses SmallRye Health to expose application health to Azure Container Apps.   The Quarkus SmallRye Health extension provides the following REST endpoints:   /q/health/live:  Indicates that the application is up and running.   /q/health/ready:  Indicates that the application is ready to accept requests.   /q/health/started:  Indicates that the application starts…  ( 49 min )

Overview This blog post shows you how to enable SmallRye Health in Quarkus applications with health probes in Azure Container Apps. The techniques shown in this blog post show you how to effectively monitor the health and status of your Quarkus instances. The application is a "to do list" with a JavaScript front end and a REST endpoint. Azure Database for PostgreSQL Flexible Server provides the persistence layer for the app. The app uses SmallRye Health to expose application health to Azure Container Apps.   The Quarkus SmallRye Health extension provides the following REST endpoints:   /q/health/live:  Indicates that the application is up and running.   /q/health/ready:  Indicates that the application is ready to accept requests.   /q/health/started:  Indicates that the application starts…

This blog post provides a step-to-step guidance for how to debug the Azure WebJobs Extension SDK both locally and remotely on the function app.   As you all know, Azure Function Apps are built on top of the Azure WebJobs SDK, which is an open-source framework that simplifies the task of writing background processing code that runs in Azure. The Azure WebJobs SDK includes a declarative binding and trigger system, commonly referred to as Azure WebJobs Extensions. These extensions allow developers to easily integrate with various Azure services and define how their background tasks respond to events.   If you encounter any issues while using the Azure WebJobs Extensions SDK, the best way to report them is via GitHub Issues. You can report bugs, request features, or ask questions there. When r…  ( 26 min )

This blog post provides a step-to-step guidance for how to debug the Azure WebJobs Extension SDK both locally and remotely on the function app.   As you all know, Azure Function Apps are built on top of the Azure WebJobs SDK, which is an open-source framework that simplifies the task of writing background processing code that runs in Azure. The Azure WebJobs SDK includes a declarative binding and trigger system, commonly referred to as Azure WebJobs Extensions. These extensions allow developers to easily integrate with various Azure services and define how their background tasks respond to events.   If you encounter any issues while using the Azure WebJobs Extensions SDK, the best way to report them is via GitHub Issues. You can report bugs, request features, or ask questions there. When r…

Modernizing Java applications and migrating to the cloud is typically a complex, labor-intensive, and fragmented process. GitHub Copilot App Modernization for Java is a powerful solution designed to simplify and accelerate your journey to the cloud. App Modernization and upgrade for Java offers an intelligent, guided approach that automates Java version upgrade and repetitive tasks and improves consistency — saving time, reducing risks, and accelerating time-to-cloud. GitHub Copilot App Modernization and upgrade for Java is in public preview and offered in a single extension pack, available in the VS Code marketplace. https://marketplace.visualstudio.com/items?itemName=vscjava.vscode-app-mod-pack   The GitHub Copilot App Modernization for Java provides six distinct value pillars, each spe…  ( 27 min )

Modernizing Java applications and migrating to the cloud is typically a complex, labor-intensive, and fragmented process. GitHub Copilot App Modernization for Java [and upgrade for Java] is a powerful solution designed to simplify and accelerate your journey to the cloud. App Modernization and upgrade for Java offers an intelligent, guided approach that automates Java version upgrade and repetitive tasks and improves consistency — saving time, reducing risks, and accelerating time-to-cloud. GitHub Copilot App Modernization and upgrade for Java is in public preview and offered in a single extension pack, available in the VS Code marketplace. https://marketplace.visualstudio.com/items?itemName=vscjava.vscode-app-mod-pack   The GitHub Copilot App Modernization for Java provides six distinct …

Generative AI is already transforming how businesses operate, with organizations seeing an average return of 3.7x for every $1 of investment [The Business Opportunity of AI, IDC study commissioned by Microsoft]. Developers sit at the center of this transformation, and their need for speed, flexibility, and familiarity with existing tools is driving the demand for application platforms that integrate AI seamlessly into their current development workflows. To fully realize the potential of generative AI in applications, organizations must provide developers with frictionless access to AI models, frameworks, and environments that enable them to scale AI applications. We see this in action at organizations like Accenture, Assembly Software, Carvana, Coldplay (Pixel Lab), Global Travel Collecti…  ( 38 min )

Generative AI is already transforming how businesses operate, with organizations seeing an average return of 3.7x for every $1 of investment [The Business Opportunity of AI, IDC study commissioned by Microsoft]. Developers sit at the center of this transformation, and their need for speed, flexibility, and familiarity with existing tools is driving the demand for application platforms that integrate AI seamlessly into their current development workflows. To fully realize the potential of generative AI in applications, organizations must provide developers with frictionless access to AI models, frameworks, and environments that enable them to scale AI applications. We see this in action at organizations like Accenture, Assembly Software, Carvana, Coldplay (Pixel Lab), Global Travel Collecti…

Azure Container Apps gives you a strong foundation for monitoring and debugging, with built-in features that give you a holistic view of your container app’s health throughout its application lifecycle. As applications grow, developers need even deeper visibility and faster ways to troubleshoot issues. That’s why we’re excited to announce new observability and debugging features. These features will help you further monitor your environment and identify root causes faster. Generally Available: OpenTelemetry agent in Azure Container Apps OpenTelemetry agent in Azure Container Apps is now generally available. This feature enables you to use open-source standards to send your app’s data without setting up the OpenTelemetry collector yourself. You can use the managed agent to choose where to s…  ( 28 min )

Azure Container Apps gives you a strong foundation for monitoring and debugging, with built-in features that give you a holistic view of your container app’s health throughout its application lifecycle. As applications grow, developers need even deeper visibility and faster ways to troubleshoot issues. That’s why we’re excited to announce new observability and debugging features. These features will help you further monitor your environment and identify root causes faster. Generally Available: OpenTelemetry agent in Azure Container Apps OpenTelemetry agent in Azure Container Apps is now generally available. This feature enables you to use open-source standards to send your app’s data without setting up the OpenTelemetry collector yourself. You can use the managed agent to choose where to s…

New App Service Premium v4 plan  The new App Service Premium v4 (Pv4) plan has entered public preview at Microsoft Build 2025 for both Windows and Linux!  This new plan is designed to support today's highly demanding application performance, scale, and budgets. Built on the latest "v6" general-purpose virtual machines and memory-optimized x64 Azure hardware with faster processors and NVMe temporary storage, it provides a noticeable performance uplift over prior generations of App Service Premium plans (over 25% in early testing). The Premium v4 offering includes nine new sizes ranging from P0v4 with a single virtual CPU and 4GB RAM all the way up through P5mv4, with 32 virtual CPUs and 256GB RAM, providing CPU and memory options to meet any business need. App Service Premium v4 plans provi…  ( 57 min )

New App Service Premium v4 plan  The new App Service Premium v4 (Pv4) plan has entered public preview at Microsoft Build 2025 for both Windows and Linux!  This new plan is designed to support today's highly demanding application performance, scale, and budgets. Built on the latest "v6" general-purpose virtual machines and memory-optimized x64 Azure hardware with faster processors and NVMe temporary storage, it provides a noticeable performance uplift over prior generations of App Service Premium plans (over 25% in early testing). The Premium v4 offering includes nine new sizes ranging from P0v4 with a single virtual CPU and 4GB RAM all the way up through P5mv4, with 32 virtual CPUs and 256GB RAM, providing CPU and memory options to meet any business need. App Service Premium v4 plans provi…

The rapid rise of AI has unlocked powerful new scenarios—from AI-powered chatbots and image generation to advanced agents. However, deploying AI models at scale presents real challenges including managing compute-intensive workloads, complexities with model deployment, and executing untrusted AI-generated code safely. Azure Container Apps addresses these challenges by offering a fully managed, flexible, serverless container platform designed for modern, cloud-native applications – now with GA of Dedicated GPUs, improved integrations for deploying Foundry models to Azure Container Apps, and the private preview of GPU powered dynamic sessions. Foundry Model Integration Azure AI Foundry Models support a wide collection of ready-to-deploy models. Traditionally, these models can be deployed wi…  ( 26 min )

The rapid rise of AI has unlocked powerful new scenarios—from AI-powered chatbots and image generation to advanced agents. However, deploying AI models at scale presents real challenges including managing compute-intensive workloads, complexities with model deployment, and executing untrusted AI-generated code safely. Azure Container Apps addresses these challenges by offering a fully managed, flexible, serverless container platform designed for modern, cloud-native applications – now with GA of Dedicated GPUs, improved integrations for deploying Foundry models to Azure Container Apps, and the private preview of GPU powered dynamic sessions. Foundry Model Integration Azure AI Foundry Models support a wide collection of ready-to-deploy models. Traditionally, these models can be deployed wi…

Azure App Service Premium v4 plan is the latest offering in the Azure App Service family, designed to deliver enhanced performance, scalability, and cost efficiency. We are excited to announce the public preview of this major upgrade to one of our most popular services. Key benefits: Fully managed platform-as-a-service (PaaS) to run your favorite web stack, on both Windows and Linux. Built using next-gen Azure hardware for higher performance and reliability. Lower total cost of ownership with new pricing tailored for large-scale app modernization projects. and more to come! Fully managed platform-as-a-service (PaaS) As the next generation of one of the leading PaaS solutions, Premium v4 abstracts infrastructure management, allowing businesses to focus on application development rather th…  ( 26 min )

Azure App Service Premium v4 plan is the latest offering in the Azure App Service family, designed to deliver enhanced performance, scalability, and cost efficiency. We are excited to announce the public preview of this major upgrade to one of our most popular services. Key benefits: Fully managed platform-as-a-service (PaaS) to run your favorite web stack, on both Windows and Linux. Built using next-gen Azure hardware for higher performance and reliability. Lower total cost of ownership with new pricing tailored for large-scale app modernization projects. and more to come! Fully managed platform-as-a-service (PaaS) As the next generation of one of the leading PaaS solutions, Premium v4 abstracts infrastructure management, allowing businesses to focus on application development rather th…

Azure Functions – Build 2025 update With Microsoft Build underway, the team is excited to provide an update on the latest releases in Azure Functions this year. Customers are leveraging Azure Functions to build AI solutions, thanks to its serverless capabilities that scale on demand and its native integration for processing real-time data. The newly launched capabilities enable the creation of AI and agentic applications with enhanced offerings, built-in security, and a pay-as-you-go model. Real-time retrieval augmented generation, making organizational data accessible through semantic search Native event driven tool function calling with the AI Foundry Agent service Hosted Model Context Protocol servers. Support for Flex consumption plans, including zone redundancy, increased regions, an…  ( 51 min )

Azure Functions – Build 2025 update With Microsoft Build underway, the team is excited to provide an update on the latest releases in Azure Functions this year. Customers are leveraging Azure Functions to build AI solutions, thanks to its serverless capabilities that scale on demand and its native integration for processing real-time data. The newly launched capabilities enable the creation of AI and agentic applications with enhanced offerings, built-in security, and a pay-as-you-go model. Real-time retrieval augmented generation, making organizational data accessible through semantic search Native event driven tool function calling with the AI Foundry Agent service Hosted Model Context Protocol servers. Support for Flex consumption plans, including zone redundancy, increased regions, an…

Building AI applications has never been more accessible. With advancements in tools and platforms, developers can now create sophisticated AI solutions that drive innovation and efficiency across various industries. For many, Kubernetes stands out as natural choice for running AI applications and agents due to its robust orchestration capabilities, scalability, and flexibility.  In this blog, we will explore the latest advancements in Azure Kubernetes Service (AKS) we are announcing at Microsoft Build 2025, designed to enhance flexibility, bolster security, and seamlessly integrate AI capabilities into your Kubernetes environments. These updates will empower developers to create sophisticated AI solutions, improve operational efficiency, and drive innovation across various industries. Let'…  ( 38 min )

Building AI applications has never been more accessible. With advancements in tools and platforms, developers can now create sophisticated AI solutions that drive innovation and efficiency across various industries. For many, Kubernetes stands out as natural choice for running AI applications and agents due to its robust orchestration capabilities, scalability, and flexibility.  In this blog, we will explore the latest advancements in Azure Kubernetes Service (AKS) we are announcing at Microsoft Build 2025, designed to enhance flexibility, bolster security, and seamlessly integrate AI capabilities into your Kubernetes environments. These updates will empower developers to create sophisticated AI solutions, improve operational efficiency, and drive innovation across various industries. Let'…

We are thrilled to announce the durable workflow capabilities in Azure Container Apps with the Durable task scheduler (preview). This new feature brings powerful workflow capabilities to Azure Container Apps, enabling developers to build and manage complex, durable workflows as code with ease. What is Workflow and the Durable task scheduler? If you’ve missed the initial announcement of the durable task scheduler, please see these existing blog posts: https://aka.ms/dts-early-access https://aka.ms/dts-public-preview In summary, the Durable task scheduler is a fully managed backend for durable execution. Durable Execution is a fault-tolerant approach to running code, designed to handle failures gracefully through automatic retries and state persistence. It is built on three core principle…  ( 26 min )

Azure Container Apps is a fully managed serverless container service that runs microservices and containerized applications on Azure. It provides built-in autoscaling, including scale to zero, and offers simplified developer experience with support for multiple programming languages and frameworks, including special features built for .NET and Java. Container Apps also provides many advanced networking and monitoring capabilities, offering seamless deployment and management of containerized applications without the need to manage underlying infrastructure. Following the features announced at Ignite’24, we've continued to innovate and enhance Azure Container Apps. We announced the general availability of Serverless GPUs, enabling seamless AI workloads with automatic scaling, optimized cold …  ( 37 min )

Durable Execution Durable Execution is a reliable approach to running code, designed to handle failures smoothly with automatic retries and state persistence. It is built on three core principles: Incremental Execution: Each operation runs independently and in order. State Persistence: The output of each step is durably saved to ensure progress is not lost. Fault Tolerance: If a step fails, the operation is retried from the last successful step, skipping previously completed steps. Durable Execution is particularly beneficial for scenarios requiring stateful chaining of operations, such as order-processing applications, data processing pipelines, ETL (extract, transform, load), and as we'll get into in this post, intelligent applications with AI agents. Durable execution simplifies the i…  ( 32 min )

As a business built by developers, for developers, Microsoft has spent decades making it faster, easier and more exciting to create great software. And developers everywhere have turned everything from BASIC and the .NET Framework, to Azure, VS Code, GitHub and more into the digital world we all live in today. But nothing compares to what’s on the horizon as agentic AI redefines both how we build and the apps we’re building. In fact, the promise of agentic AI is so strong that market forecasts predict we’re on track to reach 1.3 billion AI Agents by 2028. Our own data, from 1,500 organizations around the world, shows agent capabilities have jumped as a driver for AI applications from near last to a top three priority when comparing deployments earlier this year to applications being define…  ( 42 min )

New Networking Capabilities in Azure Container Apps Azure Container Apps is your go-to fully managed serverless container service that enables you to deploy and run containerized applications with per-second billing and autoscaling without having to manage infrastructure.   Today, Azure Container Apps is thrilled to announce several new enterprise capabilities that will take the flexibility, security, and manageability of your containerized applications to the next level. These capabilities include premium ingress, rule-based routing, private endpoints, Azure Arc integration, and planned maintenance. Let’s dive into the advanced networking features that Azure Container Apps has introduced. Public Preview: Premium Ingress in Azure Container Apps Azure Container Apps now supports premium ing…  ( 24 min )

If you’ve ever been to Microsoft Build, you know it’s not just another tech conference, it’s where ideas spark, connections happen, and the future of software takes shape. This year in Seattle, the energy is off the charts. And for those of us passionate about app modernization, it’s a moment we’ve been building toward (pun intended). At Microsoft, we believe modernization is more than just updating old systems—it’s about unlocking new possibilities with AI at the center. And at Build 2025, we’re showing exactly how we’re doing that. AI Is Changing Everything—And We’re Here for It Let’s start with the big picture: AI is transforming the entire software development lifecycle. From how we build apps to how we manage and scale them, AI is the force multiplier, reshaping how we work, build, de…  ( 35 min )

We are thrilled to announce the durable workflow capabilities in Azure Container Apps with the Durable task scheduler (preview). This new feature brings powerful workflow capabilities to Azure Container Apps, enabling developers to build and manage complex, durable workflows as code with ease. What is Workflow and the Durable task scheduler? If you’ve missed the initial announcement of the durable task scheduler, please see these existing blog posts: https://aka.ms/dts-early-access https://aka.ms/dts-public-preview In summary, the Durable task scheduler is a fully managed backend for durable execution. Durable Execution is a fault-tolerant approach to running code, designed to handle failures gracefully through automatic retries and state persistence. It is built on three core principle…

Azure Container Apps is a fully managed serverless container service that runs microservices and containerized applications on Azure. It provides built-in autoscaling, including scale to zero, and offers simplified developer experience with support for multiple programming languages and frameworks, including special features built for .NET and Java. Container Apps also provides many advanced networking and monitoring capabilities, offering seamless deployment and management of containerized applications without the need to manage underlying infrastructure. Following the features announced at Ignite’24, we've continued to innovate and enhance Azure Container Apps. We announced the general availability of Serverless GPUs, enabling seamless AI workloads with automatic scaling, optimized cold …

Durable Execution Durable Execution is a reliable approach to running code, designed to handle failures smoothly with automatic retries and state persistence. It is built on three core principles: Incremental Execution: Each operation runs independently and in order. State Persistence: The output of each step is durably saved to ensure progress is not lost. Fault Tolerance: If a step fails, the operation is retried from the last successful step, skipping previously completed steps. Durable Execution is particularly beneficial for scenarios requiring stateful chaining of operations, such as order-processing applications, data processing pipelines, ETL (extract, transform, load), and as we'll get into in this post, intelligent applications with AI agents. Durable execution simplifies the i…

As a business built by developers, for developers, Microsoft has spent decades making it faster, easier and more exciting to create great software. And developers everywhere have turned everything from BASIC and the .NET Framework, to Azure, VS Code, GitHub and more into the digital world we all live in today. But nothing compares to what’s on the horizon as agentic AI redefines both how we build and the apps we’re building. In fact, the promise of agentic AI is so strong that market forecasts predict we’re on track to reach 1.3 billion AI Agents by 2028. Our own data, from 1,500 organizations around the world, shows agent capabilities have jumped as a driver for AI applications from near last to a top three priority when comparing deployments earlier this year to applications being define…

New Networking Capabilities in Azure Container Apps Azure Container Apps is your go-to fully managed serverless container service that enables you to deploy and run containerized applications with per-second billing and autoscaling without having to manage infrastructure.   Today, Azure Container Apps is thrilled to announce several new enterprise capabilities that will take the flexibility, security, and manageability of your containerized applications to the next level. These capabilities include premium ingress, rule-based routing, private endpoints, Azure Arc integration, and planned maintenance. Let’s dive into the advanced networking features that Azure Container Apps has introduced. Public Preview: Premium Ingress in Azure Container Apps Azure Container Apps now supports premium ing…

If you’ve ever been to Microsoft Build, you know it’s not just another tech conference, it’s where ideas spark, connections happen, and the future of software takes shape. This year in Seattle, the energy is off the charts. And for those of us passionate about app modernization, it’s a moment we’ve been building toward (pun intended). At Microsoft, we believe modernization is more than just updating old systems—it’s about unlocking new possibilities with AI at the center. And at Build 2025, we’re showing exactly how we’re doing that. AI Is Changing Everything—And We’re Here for It Let’s start with the big picture: AI is transforming the entire software development lifecycle. From how we build apps to how we manage and scale them, AI is the force multiplier, reshaping how we work, build, de…

Discover powerful new features in Azure Event Grid, enhancing its functionality and user experience. This fully managed event broker now supports multi-protocol interoperability, including MQTT, for scalable messaging. It seamlessly connects Microsoft-native and third-party services, enabling robust event-driven applications. Streamline event management with flexible push-pull communication patterns.   We are thrilled to announce General Availability of the Cross-tenant delivery to Event Hubs, Service Bus, Storage Queues, and dead letter storage using managed identity with federated identity credentials (FIC) from Azure Event Grid topics, domains, system topics, and partner topics. New cross-tenant scenarios, currently in Public Preview enable delivery to Event Hubs, webhooks, and dead let…  ( 23 min )

Discover powerful new features in Azure Event Grid, enhancing its functionality and user experience. This fully managed event broker now supports multi-protocol interoperability, including MQTT, for scalable messaging. It seamlessly connects Microsoft-native and third-party services, enabling robust event-driven applications. Streamline event management with flexible push-pull communication patterns.   We are thrilled to announce General Availability of the Cross-tenant delivery to Event Hubs, Service Bus, Storage Queues, and dead letter storage using managed identity with federated identity credentials (FIC) from Azure Event Grid topics, domains, system topics, and partner topics. New cross-tenant scenarios, currently in Public Preview enable delivery to Event Hubs, webhooks, and dead let…

At Build 2025, we are excited to announce new features in Azure Monitor designed to enhance observability for developers and SREs, making it easier for you to streamline troubleshooting, improve monitoring efficiency, and gain deeper insights into application performance. With our new AI-powered tools, customizable alerts, and advanced visualization capabilities, we’re empowering developers to deliver high-quality, resilient applications with greater operational efficiency. AI-Powered Troubleshooting Capabilities We are excited to disclose two new AI-powered features, as well as share an update to a GA feature, which enhance troubleshooting and monitoring: AI-powered investigations (Public Preview): Identifies possible explanations for service degradations via automated analyses, consolid…  ( 29 min )

At Build 2025, we are excited to announce new features in Azure Monitor designed to enhance observability for developers and SREs, making it easier for you to streamline troubleshooting, improve monitoring efficiency, and gain deeper insights into application performance. With our new AI-powered tools, customizable alerts, and advanced visualization capabilities, we’re empowering developers to deliver high-quality, resilient applications with greater operational efficiency. AI-Powered Troubleshooting Capabilities We are excited to disclose two new AI-powered features, as well as share an update to a GA feature, which enhance troubleshooting and monitoring: AI-powered investigations (Public Preview): Identifies possible explanations for service degradations via automated analyses, consolid…

Troubleshooting modern cloud-native workloads has become increasingly complex. As applications scale across distributed services and regions, pinpointing the root cause of performance degradation or outages often requires navigating a maze of disconnected signals, metrics, and alerts. This fragmented experience slows down troubleshooting and burdens engineering teams with manual correlation work.  We address these challenges by introducing a unified, intelligent concept of workload health that’s enriched with application context. Health models streamline how you monitor, assess, and respond to issues affecting your workloads. Built on Azure Service Groups, they provide an out-of-the-box model tailored to your environment, consolidate signals to reduce alert noise, and surface actionable in…  ( 29 min )

Troubleshooting modern cloud-native workloads has become increasingly complex. As applications scale across distributed services and regions, pinpointing the root cause of performance degradation or outages often requires navigating a maze of disconnected signals, metrics, and alerts. This fragmented experience slows down troubleshooting and burdens engineering teams with manual correlation work.  We address these challenges by introducing a unified, intelligent concept of workload health that’s enriched with application context. Health models streamline how you monitor, assess, and respond to issues affecting your workloads. Built on Azure Service Groups, they provide an out-of-the-box model tailored to your environment, consolidate signals to reduce alert noise, and surface actionable in…

In line with our commitment to open-source solutions, we are announcing the public preview of Azure Monitor dashboards with Grafana. This service offers a powerful solution for cloud-native monitoring and visualizing Prometheus metrics. Dashboards with Grafana enable you to create and edit Grafana dashboards directly in the Azure portal without additional cost and less administrative overhead compared to self-hosting Grafana or using managed Grafana services. Start quickly with pre-built and community dashboards Pre-built Grafana dashboards for Azure Kubernetes Services, Azure Monitor, and dozens of other Azure resources are included and enabled by default. Additionally, you can import dashboards from thousands of publicly available Grafana community and open-source dashboards for Prometh…  ( 22 min )

In line with our commitment to open-source solutions, we are announcing the public preview of Azure Monitor dashboards with Grafana. This service offers a powerful solution for cloud-native monitoring and visualizing Prometheus metrics. Dashboards with Grafana enable you to create and edit Grafana dashboards directly in the Azure portal without additional cost and less administrative overhead compared to self-hosting Grafana or using managed Grafana services. Start quickly with pre-built and community dashboards Pre-built Grafana dashboards for Azure Kubernetes Services, Azure Monitor, and dozens of other Azure resources are included and enabled by default. Additionally, you can import dashboards from thousands of publicly available Grafana community and open-source dashboards for Prometh…

Public Preview: Simple Log Alerts in Azure Monitor We are excited to announce the Public Preview of Simple Log Alerts in Azure Monitor, available starting in mid-May. This new feature is designed to provide a simplified and more intuitive experience for monitoring and alerting, enhancing your ability to detect and respond to issues in near real-time. Simple Log Alerts are a new type of Log Search Alerts in Azure Monitor, designed to provide a simpler and faster alternative to Log Search Alerts. Unlike Log Search Alerts that aggregate rows over a defined period, Simple Log Alerts evaluate each row individually. This feature is now available for customers using Basic Logs who want to enable alerting. Previously, when customers opted to configure the traces table in Azure Monitor Application …  ( 21 min )

We’re thrilled to announce the general availability (GA) of Managed Prometheus visualizations in Azure Monitor for AKS, along with an enhanced, unified AKS Monitoring experience. Troubleshooting Kubernetes clusters is often time-consuming and complex whether you're diagnosing failures, scaling issues, or performance bottlenecks. This redesign of the existing Insights experience brings all your key monitoring data into a single, streamlined view reducing the time and effort it takes to diagnose, triage, and resolve problems so you can keep your applications running smoothly with less manual work. By using Managed Prometheus, customers can also realize up to 80% savings on metrics costs and benefit from up to 90% faster blade load performance delivering both a powerful and cost-efficient way…  ( 25 min )

Investigate smarter – click, analyze, and easily mitigate with Azure Monitor investigations! We are excited to introduce the public preview of Azure Monitor issue and investigation.      These new capabilities are designed to enhance your troubleshooting experience and streamline the process of resolving health degradations in your application and infrastructure. What it is  Azure Monitor investigation is an AI-powered, automated analysis designed to scan the telemetry gathered by Azure Monitor to troubleshoot and mitigate potential service health issues. Investigation provides a list of AI-generated findings that include a summary of what happened, potential causes, and steps for further troubleshooting and mitigation. Azure Monitor issue contains all observability related data and proce…  ( 23 min )

We are thrilled to announce the launch of a new feature in Azure Monitor: Customizable Email Subjects for Log Search Alerts V2, available during May. What it is Customizable Email Subjects for Log Search Alerts V2 is a new feature that enables customers to personalize the subject lines of alert emails, making it easier to quickly identify and respond to alerts with more relevant and specific information. How it works This feature allows you to override email subjects with dynamic values by concatenating information from the common schema and custom text. For example, you can customize email subjects to include specific details such as the name of the virtual machine (VM) or patching details, allowing for quick identification without opening the email. Getting Started To get started with Cu…  ( 22 min )

Public Preview: Simple Log Alerts in Azure Monitor We are excited to announce the Public Preview of Simple Log Alerts in Azure Monitor, available starting in mid-May. This new feature is designed to provide a simplified and more intuitive experience for monitoring and alerting, enhancing your ability to detect and respond to issues in near real-time. Simple Log Alerts are a new type of Log Search Alerts in Azure Monitor, designed to provide a simpler and faster alternative to Log Search Alerts. Unlike Log Search Alerts that aggregate rows over a defined period, Simple Log Alerts evaluate each row individually. This feature is now available for customers using Basic Logs who want to enable alerting. Previously, when customers opted to configure the traces table in Azure Monitor Application …

We’re thrilled to announce the general availability (GA) of Managed Prometheus visualizations in Azure Monitor for AKS, along with an enhanced, unified AKS Monitoring experience. Troubleshooting Kubernetes clusters is often time-consuming and complex whether you're diagnosing failures, scaling issues, or performance bottlenecks. This redesign of the existing Insights experience brings all your key monitoring data into a single, streamlined view reducing the time and effort it takes to diagnose, triage, and resolve problems so you can keep your applications running smoothly with less manual work. By using Managed Prometheus, customers can also realize up to 80% savings on metrics costs and benefit from up to 90% faster blade load performance delivering both a powerful and cost-efficient way…

Investigate smarter – click, analyze, and easily mitigate with Azure Monitor investigations! We are excited to introduce the public preview of Azure Monitor issue and investigation.      These new capabilities are designed to enhance your troubleshooting experience and streamline the process of resolving health degradations in your application and infrastructure. What it is  Azure Monitor investigation is an AI-powered, automated analysis designed to scan the telemetry gathered by Azure Monitor to troubleshoot and mitigate potential service health issues. Investigation provides a list of AI-generated findings that include a summary of what happened, potential causes, and steps for further troubleshooting and mitigation. Azure Monitor issue contains all observability related data and proce…

We are thrilled to announce the launch of a new feature in Azure Monitor: Customizable Email Subjects for Log Search Alerts V2, available during May. What it is Customizable Email Subjects for Log Search Alerts V2 is a new feature that enables customers to personalize the subject lines of alert emails, making it easier to quickly identify and respond to alerts with more relevant and specific information. How it works This feature allows you to override email subjects with dynamic values by concatenating information from the common schema and custom text. For example, you can customize email subjects to include specific details such as the name of the virtual machine (VM) or patching details, allowing for quick identification without opening the email. Getting Started To get started with Cu…

AI is reshaping how we build, deploy, and scale software. As more apps become AI-powered, developers need environments that can keep up with speed and demands of innovation. We’ve heard from customers just how critical it is to have a zero-config experience when building AI applications, with streamlined access to compute, prebuilt models, and environment […] The post Supercharge AI development with new AI-powered features in Microsoft Dev Box appeared first on Develop from the cloud.  ( 32 min )

AI agents, model context protocols (MCPs), and emerging AI workflows are fundamentally transforming software development paradigms and broadening the range of applications we can create. Developers are expected to keep up with the pace of innovation, but that’s hard to do when working from traditional development environments or legacy VDI solutions. With Microsoft Dev Box, […] The post Unlock developer potential with Microsoft Dev Box appeared first on Develop from the cloud.  ( 27 min )

Learn how Microsoft 365 Copilot APIs allow you to build solutions grounded in your organization’s content, context, and permissions, without needing to relocate or duplicate data. The post Introducing Microsoft 365 Copilot APIs appeared first on Microsoft 365 Developer Blog.  ( 23 min )

Today we’re thrilled to introduce Azure SRE Agent, an AI-powered tool that makes it easier to sustain production cloud environments. SRE Agent helps respond to incidents quickly and effectively, alleviating the toil of managing production environments. Overall, it results in better service uptime and reduced operational costs. SRE Agent leverages the reasoning capabilities of LLMs to identify the logs and metrics necessary for rapid root cause analysis and issue mitigation. Its advanced AI capabilities transform incident and infrastructure management in Azure, freeing engineers to focus on more meaningful work.  Sign up for the SRE Agent preview click here As more companies move their services online, site reliability engineering (SRE) has become crucial to keeping critical systems reliab…  ( 28 min )

Today we’re thrilled to introduce Azure SRE Agent, an AI-powered tool that makes it easier to sustain production cloud environments. SRE Agent helps respond to incidents quickly and effectively, alleviating the toil of managing production environments. Overall, it results in better service uptime and reduced operational costs. SRE Agent leverages the reasoning capabilities of LLMs to identify the logs and metrics necessary for rapid root cause analysis and issue mitigation. Its advanced AI capabilities transform incident and infrastructure management in Azure, freeing engineers to focus on more meaningful work.  Sign up for the SRE Agent preview click here As more companies move their services online, site reliability engineering (SRE) has become crucial to keeping critical systems reliab…

Modernizing Java applications and migrating to the cloud is often a complex, time-consuming, and fragmented process. GitHub Copilot app modernization for Java [and upgrade for Java] is a powerful solution designed to simplify and accelerate your journey to the cloud. Available now in Public Preview as a single extension pack in the Visual Studio Code Marketplace, […] The post Announcing Public Preview of the GitHub Copilot app modernization for Java appeared first on Microsoft for Java Developers.  ( 24 min )

Introduction  We're thrilled to introduce a new suite of multimodal capabilities in Azure AI Search.     This set of features includes both new additions and incremental improvements that enable Azure AI Search to extract text from pages and inline images, generate image descriptions (verbalization), and create vision/text embeddings. It also facilitates storing cropped images in the knowledge store and returning text and image annotations in RAG (Retrieval Augmented Generation) applications to end users.  These features can be configured in our new Azure portal wizard with multimodal support or via the REST API 2025-05-01-preview version.  In addition, we're providing a new GitHub repo with sample code for a RAG app. This resource shows how you can use the index created by Azure AI Search…  ( 34 min )

Today we’re announcing agentic retrieval in Azure AI Search, a multiturn query engine that plans and runs its own retrieval strategy for improved answer relevance. Compared to traditional, single-shot RAG, agentic retrieval improves answer relevance to complex questions by up to 40%.  It transforms queries, runs parallel searches, and delivers results tuned for agents, along with references and a query activity log. Now available in public preview. What is agentic retrieval? Agentic retrieval in Azure AI Search uses a new query architecture that incorporates user conversation history and an Azure OpenAI model to plan, retrieve and synthesize queries. Here's how it works: An LLM analyzes the entire chat thread to identify the underlying information need. Instead of a single, catch-all que…  ( 26 min )

By Alec Berntson, Alina Stoica Beck, Amaia Salvador Aguilera, Arnau Quindós Sánchez, Thibault Gisselbrecht and Xianshun Chen   Agentic retrieval in Azure AI Search is a new API built to effectively answer complex queries by extracting the right content needed. The API defines and runs a query plan, incorporating conversation history and an Azure OpenAI model. It transforms complex queries, then performs multiple searches at once, combining the final results and delivering ready-to-use content for answer generation. In this post we detail the operations that take place while the API is called and walk through the numerous experiments and datasets to evaluate its relevance performance. We learned the agentic retrieval API automates optimal retrieval for complex user queries, so you can get m…  ( 74 min )

Introduction  We're thrilled to introduce a new suite of multimodal capabilities in Azure AI Search.     This set of features includes both new additions and incremental improvements that enable Azure AI Search to extract text from pages and inline images, generate image descriptions (verbalization), and create vision/text embeddings. It also facilitates storing cropped images in the knowledge store and returning text and image annotations in RAG (Retrieval Augmented Generation) applications to end users.  These features can be configured in our new Azure portal wizard with multimodal support or via the REST API 2025-05-01-preview version.  In addition, we're providing a new GitHub repo with sample code for a RAG app. This resource shows how you can use the index created by Azure AI Search…

Today we’re announcing agentic retrieval in Azure AI Search, a multiturn query engine that plans and runs its own retrieval strategy for improved answer relevance. Compared to traditional, single-shot RAG, agentic retrieval improves answer relevance to complex questions by up to 40%.  It transforms queries, runs parallel searches, and delivers results tuned for agents, along with references and a query activity log. Now available in public preview. What is agentic retrieval? Agentic retrieval in Azure AI Search uses a new query architecture that incorporates user conversation history and an Azure OpenAI model to plan, retrieve and synthesize queries. Here's how it works: An LLM analyzes the entire chat thread to identify the underlying information need. Instead of a single, catch-all que…

By Alec Berntson, Alina Stoica Beck, Amaia Salvador Aguilera, Arnau Quindós Sánchez, Thibault Gisselbrecht and Xianshun Chen   Agentic retrieval in Azure AI Search is a new API built to effectively answer complex queries by extracting the right content needed. The API defines and runs a query plan, incorporating conversation history and an Azure OpenAI model. It transforms complex queries, then performs multiple searches at once, combining the final results and delivering ready-to-use content for answer generation. In this post we detail the operations that take place while the API is called and walk through the numerous experiments and datasets to evaluate its relevance performance. We learned the agentic retrieval API automates optimal retrieval for complex user queries, so you can get m…

The Fabric Spark Native Execution Engine (NEE) is now generally available (GA) as part of Fabric Runtime 1.3. This C++-based vectorized engine (built on Apache Gluten and Velox) runs Spark workloads directly on the lakehouse, requiring no code changes or new libraries. It supports Spark 3.5 APIs and both Parquet and Delta Lake formats, so … Continue reading “Microsoft Fabric Spark: Native Execution Engine now generally available”  ( 6 min )

PostgreSQL is a popular relational database in application development, and Azure Database for PostgreSQL provides a fully managed service with enterprise-level security, availability, and scalability. Integrating Azure Database for PostgreSQL with Microsoft Fabric through Mirroring (Preview) enables seamless replication of transactional data for analytics, simplifying data processes and ensuring real-time insights and data consistency. Microsoft … Continue reading “Simplify Your Data Strategy: Mirroring for Azure Database for PostgreSQL in Microsoft Fabric for Effortless Analytics on Transactional Data”  ( 8 min )

Mirroring in Microsoft Fabric allows you to seamlessly reflect your existing data estate from Azure SQL Managed Instance into OneLake. The mirrored data is automatically kept up to date in near real-time, enabling advanced analytics and reporting to generate essential business insights. Since our preview announcement of Mirroring for Azure SQL Managed Instance, we have listened … Continue reading “New features in Mirroring for Azure SQL Managed Instance – private endpoint support and more”  ( 6 min )

At Microsoft Build 2025, we are thrilled to show you the latest innovations that we have delivered with Mirroring in Fabric. Mirroring is a powerful feature that allows you to seamlessly reflect your existing data estate continuously from any database or data warehouse into OneLake in Fabric. Once Mirroring starts the replication process, the mirrored … Continue reading “What’s new with Mirroring in Fabric at Microsoft Build 2025”  ( 9 min )

We’re excited to announce the General Availability of Dataflow Gen2 CI/CD & Git integration support! With this set of features, you can now seamlessly integrate your Dataflow Gen2 items with your existing CI/CD pipelines and version control of your workspace in Fabric. This integration allows for better collaboration, versioning, and automation of your deployment process … Continue reading “Dataflow Gen2 CI/CD, GIT integration and Public APIs (Generally Available)”  ( 7 min )

As organizations advance in their cloud platform journey, ensuring robust data security remains fundamental. Encryption plays a crucial role in defense-in-depth strategies used to safeguard sensitive information by adding a layer of protection against unauthorized access. In addition to strengthening your security posture, encryption helps you adhere to your organization’s internal security, data governance and … Continue reading “Encrypt data at rest in your Fabric workspaces using customer-managed keys (Preview)”  ( 7 min )

Maintaining data consistency during ETL (Extract, Transform, Load) processes has long been a critical challenge for data engineers. Whether it’s a nightly pipeline overwriting key records or a mid-day transformation introducing schema drift, the risk of disrupting downstream analytics is both real and costly. In today’s fast-paced, data-driven world, even brief inconsistencies can break dashboards, … Continue reading “Warehouse Snapshots in Microsoft Fabric (Preview)”  ( 11 min )

A recent study from the Social Science Research Network looked at 5,000 developers using generative AI tools in their day-to-day work and found a 26% average increase in completed tasks. The massive opportunity generative AI presents for developers and data professionals was one of the key driving forces behind the initial development of Microsoft Fabric. … Continue reading “Get to insights faster with SaaS databases and “chat with your data””  ( 14 min )

We are excited to announce Materialized Lake views (MLV) in Microsoft Fabric. Coming soon in preview, MLV is a new feature that allows you to build declarative data pipelines using SQL, complete with built-in data quality rules and automatic monitoring of data transformations. In essence, an MLV is a persisted, continuously updated view of your … Continue reading “Simplifying Medallion Implementation with Materialized Lake Views in Fabric”  ( 9 min )

We’re excited to announce a new shortcut type for Azure Blob Storage in Microsoft Fabric! As a key platform for storing unstructured data — from images and documents to logs and media — Azure Blob Storage plays a vital role in powering AI and advanced analytics solutions. With this new shortcut type, you can easily … Continue reading “New Shortcut Type for Azure Blob Storage in OneLake shortcuts”  ( 6 min )

Announcing preview of Cosmos DB in Microsoft Fabric. Cosmos DB makes it easy to build AI apps, offering a database that scales automatically, deeply integrated with OneLake ad and is secure right out of the box. It’s built on scalability and performance of Azure Cosmos DB.  ( 8 min )

Control who can find, use, and create agents, define permissions, approve or block agent deployments, and configure billing models including pay-as-you-go or prepaid options. Get detailed visibility into how agents are used, which users and groups are driving consumption, and how much they’re costing you. With Microsoft Purview integration, get visibility into sensitive data exposure, track compliance risks, and audit agent activity to stay secure and aligned with your organization’s data policies. Jeremy Chapman, Director of Microsoft 365, shares how to configure, deploy, monitor, and secure AI agents at scale. Define agent access by group or user. Customize permissions with Microsoft 365 admin controls. See how to use the Copilot Control System. Enable pay-as-you-go agent billing with m…  ( 38 min )

Control who can find, use, and create agents, define permissions, approve or block agent deployments, and configure billing models including pay-as-you-go or prepaid options. Get detailed visibility into how agents are used, which users and groups are driving consumption, and how much they’re costing you. With Microsoft Purview integration, get visibility into sensitive data exposure, track compliance risks, and audit agent activity to stay secure and aligned with your organization’s data policies. Jeremy Chapman, Director of Microsoft 365, shares how to configure, deploy, monitor, and secure AI agents at scale. Define agent access by group or user. Customize permissions with Microsoft 365 admin controls. See how to use the Copilot Control System. Enable pay-as-you-go agent billing with m…

Table of Contents Abstract Introduction Enterprise RAG: Challenges and Requirements NVIDIA AI Blueprint for RAG Adapting the Blueprint for Azure Deployment Azure NetApp Files: Powering High-Performance RAG Workloads Why Azure NetApp Files works well for RAG Service Levels for RAG Workloads Dynamic Service Level Adjustments Snapshot Capabilities for ML Versioning Cost Optimization Strategies Azure Reference Architecture for Enterprise RAG End-to-End Workflow Implementation Guide: Building the Pipeline Setup your Bash shell Setup your Azure Account Set environment variables Evaluating Your Enterprise RAG Pipeline Retrieval Accuracy Latency and Throughput GPU Utilization Cost Analysis What’s Coming Next Enterprise Use Cases and Real-World Applications Enterprise Search Customer Support Regula…  ( 81 min )

Table of Contents Abstract Introduction Enterprise RAG: Challenges and Requirements NVIDIA AI Blueprint for RAG Adapting the Blueprint for Azure Deployment Azure NetApp Files: Powering High-Performance RAG Workloads Why Azure NetApp Files works well for RAG Service Levels for RAG Workloads Dynamic Service Level Adjustments Snapshot Capabilities for ML Versioning Cost Optimization Strategies Azure Reference Architecture for Enterprise RAG End-to-End Workflow Implementation Guide: Building the Pipeline Setup your Bash shell Setup your Azure Account Set environment variables Evaluating Your Enterprise RAG Pipeline Retrieval Accuracy Latency and Throughput GPU Utilization Cost Analysis What’s Coming Next Enterprise Use Cases and Real-World Applications Enterprise Search Customer Support Regula…

In the ever-evolving world of Generative AI, the pace of innovation is nothing short of breathtaking. Just a few short months ago, Large Language Models (LLMs) and their smaller counterparts, Small Language Models (SLMs), were the talk of the town. Then came Retrieval Augmented Generation (RAG), offering a powerful way to ground these models to specific knowledge bases. The emergence of Agents and Agentic AI further opened doors for new possibilities by using the GenAI Model capabilities. Now, as the next step in this exciting journey, we're witnessing the arrival of a new open protocol – one that standardizes how applications provide crucial context to LLMs. This new protocol is MCP. Model Context Protocol: Currently there is a Challenge while integrating LLMs with specific tools (like da…  ( 42 min )

In the ever-evolving world of Generative AI, the pace of innovation is nothing short of breathtaking. Just a few short months ago, Large Language Models (LLMs) and their smaller counterparts, Small Language Models (SLMs), were the talk of the town. Then came Retrieval Augmented Generation (RAG), offering a powerful way to ground these models to specific knowledge bases. The emergence of Agents and Agentic AI further opened doors for new possibilities by using the GenAI Model capabilities. Now, as the next step in this exciting journey, we're witnessing the arrival of a new open protocol – one that standardizes how applications provide crucial context to LLMs. This new protocol is MCP. Model Context Protocol: Currently there is a Challenge while integrating LLMs with specific tools (like da…

We’re laying the groundwork to bring .NET Aspire to Azure App Service. While this is just the beginning, we wanted to give you an early preview of how to set up a basic Aspire application on App Service.  ( 5 min )

Model Context Protocol is an open standard that acts as a universal connector between AI models and the outside world. Think of MCP as “the USB-C of the AI world,” allowing AI systems to plug into APIs, databases, files, and other tools seamlessly. By adopting MCP, developers can create smarter, more useful AI applications that access up-to-date information and perform actions like a human developer would. To help developers learn this game-changing technology, Microsoft has created the “MCP for Beginners” course a free, open-source curriculum that guides you from the basics of MCP to building real-world AI integrations. Below, we’ll explore what MCP is, who this course is for, and how it empowers both beginners and intermediate developers to get started with MCP. What is MCP and Why Shoul…  ( 43 min )

Model Context Protocol is an open standard that acts as a universal connector between AI models and the outside world. Think of MCP as “the USB-C of the AI world,” allowing AI systems to plug into APIs, databases, files, and other tools seamlessly. By adopting MCP, developers can create smarter, more useful AI applications that access up-to-date information and perform actions like a human developer would. To help developers learn this game-changing technology, Microsoft has created the “MCP for Beginners” course a free, open-source curriculum that guides you from the basics of MCP to building real-world AI integrations. Below, we’ll explore what MCP is, who this course is for, and how it empowers both beginners and intermediate developers to get started with MCP. What is MCP and Why Shoul…

With built-in visibility into how AI apps and agents interact with sensitive data — whether inside Microsoft 365 or across unmanaged consumer tools — you can detect risks early, take decisive action, and enforce the right protections without slowing innovation. See usage trends, investigate prompts and responses, and respond to potential data oversharing or policy violations in real time. From compliance-ready audit logs to adaptive data protection, you’ll have the insights and tools to keep data secure as AI becomes a part of everyday work. Shilpa Ranganathan, Microsoft Purview Principal Group PM, shares how to balance AI innovation with enterprise-grade data governance and security. Move from detection to prevention. Built-in, pre-configured policies you can activate in seconds. Check o…  ( 43 min )

With built-in visibility into how AI apps and agents interact with sensitive data — whether inside Microsoft 365 or across unmanaged consumer tools — you can detect risks early, take decisive action, and enforce the right protections without slowing innovation. See usage trends, investigate prompts and responses, and respond to potential data oversharing or policy violations in real time. From compliance-ready audit logs to adaptive data protection, you’ll have the insights and tools to keep data secure as AI becomes a part of everyday work. Shilpa Ranganathan, Microsoft Purview Principal Group PM, shares how to balance AI innovation with enterprise-grade data governance and security. Move from detection to prevention. Built-in, pre-configured policies you can activate in seconds. Check o…

Unify and secure your data — no matter where it lives — without sacrificing control using OneLake security, part of Microsoft Fabric.  With granular permissions down to the row, column, and table level, you can confidently manage access across engines like Power BI, Spark, and T-SQL, all from one place. Discover, label, and govern your data with clarity using the integrated OneLake catalog that surfaces the right items fast. Aaron Merrill, Microsoft Fabric Principal Program Manager, shows how you can stay in control, from security to discoverability — owning, sharing, and protecting data on your terms. Protect sensitive information at scale. Set precise data access rules — down to individual rows. Check out OneLake security in Microsoft Fabric. No data duplication needed. Hide sensitive…  ( 41 min )

Unify and secure your data — no matter where it lives — without sacrificing control using OneLake security, part of Microsoft Fabric.  With granular permissions down to the row, column, and table level, you can confidently manage access across engines like Power BI, Spark, and T-SQL, all from one place. Discover, label, and govern your data with clarity using the integrated OneLake catalog that surfaces the right items fast. Aaron Merrill, Microsoft Fabric Principal Program Manager, shows how you can stay in control, from security to discoverability — owning, sharing, and protecting data on your terms. Protect sensitive information at scale. Set precise data access rules — down to individual rows. Check out OneLake security in Microsoft Fabric. No data duplication needed. Hide sensitive…

Streamline your day with new, user-focused updates to Microsoft 365 Copilot. Jump into work faster with a redesigned layout that puts Chat, Search, and your agents front and center. New Copilot Search lets you yse natural language to find files, emails, and conversations — even if you don’t remember exact keywords — and get instant summaries and previews without switching apps. Create high-impact visuals, documents, and videos in seconds with the new Copilot Create experience, complete with support for brand templates. Tap into powerful agents like Researcher and Analyst to handle deep tasks or build your own with ease. And if you manage Copilot across your organization, you now have better tools to deploy, monitor, and secure AI use — all from a single view. Describe what you want. Don’…  ( 42 min )

Streamline your day with new, user-focused updates to Microsoft 365 Copilot. Jump into work faster with a redesigned layout that puts Chat, Search, and your agents front and center. New Copilot Search lets you yse natural language to find files, emails, and conversations — even if you don’t remember exact keywords — and get instant summaries and previews without switching apps. Create high-impact visuals, documents, and videos in seconds with the new Copilot Create experience, complete with support for brand templates. Tap into powerful agents like Researcher and Analyst to handle deep tasks or build your own with ease. And if you manage Copilot across your organization, you now have better tools to deploy, monitor, and secure AI use — all from a single view. Describe what you want. Don’…

What are Availability Zones?  Availability Zones, or zone redundancy, refers to the deployment of applications across multiple availability zones within an Azure region. Each availability zone consists of one or more data centers with independent power, cooling, and networking. By leveraging zone redundancy, you can protect your applications and data from data center failures, ensuring uninterrupted service.  Key Updates  The minimum instance requirement for enabling Availability Zones has been reduced from three instances to two, while still maintaining a 99.99% SLA.  Many existing App Service plans with two or more instances will automatically support Availability Zones without additional setup.  The zone redundant setting for App Service plans and App Service Environment v3 is now …  ( 32 min )

What are Availability Zones?  Availability Zones, or zone redundancy, refers to the deployment of applications across multiple availability zones within an Azure region. Each availability zone consists of one or more data centers with independent power, cooling, and networking. By leveraging zone redundancy, you can protect your applications and data from data center failures, ensuring uninterrupted service.  Key Updates  The minimum instance requirement for enabling Availability Zones has been reduced from three instances to two, while still maintaining a 99.99% SLA.  Many existing App Service plans with two or more instances will automatically support Availability Zones without additional setup.  The zone redundant setting for App Service plans and App Service Environment v3 is now …

It’s that time of year again—Microsoft Build 2025 is here! And in the spirit of pushing boundaries with AI, we’re thrilled to introduce a powerful new preview feature in Conversational Diagnostics. 📸 Diagnose with a ScreenshotNo more struggling to describe a tricky issue or typing out long explanations. With this new capability, you can simply paste, upload, or drag a screenshot into the chat. Conversational Diagnostics will analyze the image, identify the context, and surface relevant diagnostics for your selected Azure Resource—all in seconds. Whether you're debugging a web app or triaging a customer issue, this feature helps you move from problem to insight faster than ever. Thank you!  ( 18 min )

It’s that time of year again—Microsoft Build 2025 is here! And in the spirit of pushing boundaries with AI, we’re thrilled to introduce a powerful new preview feature in Conversational Diagnostics. 📸 Diagnose with a ScreenshotNo more struggling to describe a tricky issue or typing out long explanations. With this new capability, you can simply paste, upload, or drag a screenshot into the chat. Conversational Diagnostics will analyze the image, identify the context, and surface relevant diagnostics for your selected Azure Resource—all in seconds. Whether you're debugging a web app or triaging a customer issue, this feature helps you move from problem to insight faster than ever. Thank you!

Cost tracking is a critical aspect of cloud operations—it helps you understand not just how much you're spending, but also where that spend is going and which teams are responsible. When running a Machine Learning capability with multiple consumers across your organisation, it becomes especially challenging to attribute compute costs to the teams building and deploying models. With the extensive compute use in Machine Learning, these costs can add up quickly. In this article, we’ll explore how tools like Kubecost can help bring visibility and accountability to ML workloads. Tracking costs in Azure can mostly be done through Azure Cost Management, however when we are running these ML models as endpoints and deployments in a Kubernetes cluster, things can get a bit trickier. Azure Cost Manag…  ( 41 min )

Cost tracking is a critical aspect of cloud operations—it helps you understand not just how much you're spending, but also where that spend is going and which teams are responsible. When running a Machine Learning capability with multiple consumers across your organisation, it becomes especially challenging to attribute compute costs to the teams building and deploying models. With the extensive compute use in Machine Learning, these costs can add up quickly. In this article, we’ll explore how tools like Kubecost can help bring visibility and accountability to ML workloads. Tracking costs in Azure can mostly be done through Azure Cost Management, however when we are running these ML models as endpoints and deployments in a Kubernetes cluster, things can get a bit trickier. Azure Cost Manag…

A New Way to Host Functions on ACA With the new native hosting model, Azure Functions are now fully integrated into ACA. This means you can deploy and run your functions directly on ACA, taking full advantage of the robust app platform.  Create via Portal: Option to optimize for Azure function If you are using CLI, you can deploy Azure Functions directly onto Azure Container Apps using the Microsoft.App resource provider by setting “kind=functionapp” property on the Container App resource. Create via CLI: Set “kind=functionapp” property Please note, in the new native hosting model, Azure Functions extensions will continue to work as before. Auto-scaling will remain available. Deployments are supported through ARM templates, Bicep, Azure CLI, and the Azure portal. Monitoring using Applicat…  ( 27 min )

A New Way to Host Functions on ACA With the new native hosting model, Azure Functions are now fully integrated into ACA. This means you can deploy and run your functions directly on ACA, taking full advantage of the robust app platform.  Create via Portal: Option to optimize for Azure function If you are using CLI, you can deploy Azure Functions directly onto Azure Container Apps using the Microsoft.App resource provider by setting “kind=functionapp” property on the Container App resource. Create via CLI: Set “kind=functionapp” property Please note, in the new native hosting model, Azure Functions extensions will continue to work as before. Auto-scaling will remain available. Deployments are supported through ARM templates, Bicep, Azure CLI, and the Azure portal. Monitoring using Applicat…

Stay in control with Microsoft Defender. You can identify which AI apps and cloud services are in use across your environment, evaluate their risk levels, and allow or block them as needed — all from one place. Whether it’s a sanctioned tool or a shadow AI app, you’re equipped to set the right policies and respond fast to emerging threats.  Microsoft Defender gives you the visibility to track complex attack paths — linking signals across endpoints, identities, and cloud apps. Investigate real-time alerts, protect sensitive data from misuse in AI tools like Copilot, and enforce controls even for in-house developed apps using system prompts and Azure AI Foundry.  Rob Lefferts, Microsoft Security CVP, joins me in the Mechanics studio to share how you can safeguard your AI-powered environment…  ( 59 min )

Stay in control with Microsoft Defender. You can identify which AI apps and cloud services are in use across your environment, evaluate their risk levels, and allow or block them as needed — all from one place. Whether it’s a sanctioned tool or a shadow AI app, you’re equipped to set the right policies and respond fast to emerging threats.  Microsoft Defender gives you the visibility to track complex attack paths — linking signals across endpoints, identities, and cloud apps. Investigate real-time alerts, protect sensitive data from misuse in AI tools like Copilot, and enforce controls even for in-house developed apps using system prompts and Azure AI Foundry.  Rob Lefferts, Microsoft Security CVP, joins me in the Mechanics studio to share how you can safeguard your AI-powered environment…

Protect your Microsoft 365 data and stay in control with Microsoft 365 Backup — whether managing email, documents, or sites across Exchange, OneDrive, and SharePoint. Define exactly what you want to back up and restore precisely what you need to with speeds reaching 2TB per hour at scale. With flexible policies, dynamic rules, and recovery points up to 365 days back, you can stay resilient and ready.  In this introduction, I'll show you how to minimize disruption and keep your organization moving forward even in the event of a disaster with Microsoft 365 Backup.  Fine-tune what gets backed up.  Back up by user, site, group, or file type — to meet your exact needs. Get started with Microsoft 365 Backup.  Restore data in-place or to a new location.  Compare versions before committing. Tak…  ( 45 min )

Protect your Microsoft 365 data and stay in control with Microsoft 365 Backup — whether managing email, documents, or sites across Exchange, OneDrive, and SharePoint. Define exactly what you want to back up and restore precisely what you need to with speeds reaching 2TB per hour at scale. With flexible policies, dynamic rules, and recovery points up to 365 days back, you can stay resilient and ready.  In this introduction, I'll show you how to minimize disruption and keep your organization moving forward even in the event of a disaster with Microsoft 365 Backup.  Fine-tune what gets backed up.  Back up by user, site, group, or file type — to meet your exact needs. Get started with Microsoft 365 Backup.  Restore data in-place or to a new location.  Compare versions before committing. Tak…

This post will discuss this message that may be seen on Azure Container Apps.  ( 8 min )

Introduction In Kenya's evolving higher education landscape, access to accurate information about funding opportunities remains a critical challenge for students and their families. When the New Funding Model (NFM) was introduced in May 2023, many applicants found themselves navigating unfamiliar processes with limited guidance. This knowledge gap inspired our team to develop a solution that would bridge this divide using cutting-edge AI technology. Our project, RAG-Powered Virtual Assistant for the Higher Education Fund (HEF), emerged as the overall winner at the Microsoft Data + AI Hack Kenya 2025. By leveraging Microsoft Fabric's powerful Eventhouse capabilities alongside Azure OpenAI services, we created an intelligent assistant that provides instant, accurate responses to queries abou…  ( 40 min )

Introduction In Kenya's evolving higher education landscape, access to accurate information about funding opportunities remains a critical challenge for students and their families. When the New Funding Model (NFM) was introduced in May 2023, many applicants found themselves navigating unfamiliar processes with limited guidance. This knowledge gap inspired our team to develop a solution that would bridge this divide using cutting-edge AI technology. Our project, RAG-Powered Virtual Assistant for the Higher Education Fund (HEF), emerged as the overall winner at the Microsoft Data + AI Hack Kenya 2025. By leveraging Microsoft Fabric's powerful Eventhouse capabilities alongside Azure OpenAI services, we created an intelligent assistant that provides instant, accurate responses to queries abou…

Introduction Ever found yourself wishing your web interface could really talk and listen back to you? With a few clicks (and a bit of code), you can turn your plain Open WebUI into a full-on voice assistant. In this post, you’ll see how to spin up an Azure Speech resource, hook it into your frontend, and watch as user speech transforms into text and your app’s responses leap off the screen in a human-like voice. By the end of this guide, you’ll have a voice-enabled web UI that actually converses with users, opening the door to hands-free controls, better accessibility, and a genuinely richer user experience. Ready to make your web app speak? Let’s dive in. Why Azure AI Speech? We use Azure AI Speech service in Open Web UI to enable voice interactions directly within web applications. This …  ( 27 min )

Introduction Ever found yourself wishing your web interface could really talk and listen back to you? With a few clicks (and a bit of code), you can turn your plain Open WebUI into a full-on voice assistant. In this post, you’ll see how to spin up an Azure Speech resource, hook it into your frontend, and watch as user speech transforms into text and your app’s responses leap off the screen in a human-like voice. By the end of this guide, you’ll have a voice-enabled web UI that actually converses with users, opening the door to hands-free controls, better accessibility, and a genuinely richer user experience. Ready to make your web app speak? Let’s dive in. Why Azure AI Speech? We use Azure AI Speech service in Open Web UI to enable voice interactions directly within web applications. This …

Semantic Kernel: Package Previews, Graduations & Deprecations We are excited to share a summary of recent updates and continuous clean-up efforts across the Semantic Kernel .NET codebase. These changes focus on improving maintainability, aligning with the latest APIs, and ensuring a consistent experience for users. Below you’ll find details on package graduations, deprecations, and a […] The post Semantic Kernel: Package previews, Graduations & Deprecations appeared first on Semantic Kernel.  ( 23 min )

Overview Microsoft Fabric API for GraphQL makes it easy to query and mutate data from a Fabric- SQL database and other Fabric data sources such as Data Warehouse and Lakehouse, with strongly typed schemas and a rich query language allowing developers to create an intuitive API without writing custom server code. While you can’t customize … Continue reading “Smart Mutations in Microsoft Fabric API for GraphQL with Stored Procedures”  ( 7 min )

Fabric Copilot Capacities are making changes to be more streamlined and easier to use.  ( 6 min )

We are excited to introduce Copilot Inline Code Completion, an AI-powered feature that helps data scientists and engineers write high-quality Python code faster and with greater ease. Inspired by GitHub Copilot, this feature offers intelligent code suggestions as you type, with no commands needed. By understanding the context of your notebook, Copilot Inline Code Completion … Continue reading “Improving productivity in Fabric Notebooks with Inline Code Completion”  ( 6 min )

If you've been curious about how to build your own AI agents that can talk to APIs, connect with tools like databases, or even follow documentation you're in the right place. Microsoft has created something called MCP, which stands for Model‑Context‑Protocol. And to help you learn it step by step, they’ve made an amazing MCP Resources Hub on GitHub. In this blog, I’ll Walk you through what MCP is, why it matters, and how to use this hub to get started, even if you're new to AI development. What is MCP (Model‑Context‑Protocol)? Think of MCP like a communication bridge between your AI model and the outside world. Normally, when we chat with AI (like ChatGPT), it only knows what’s in its training data. But with MCP, you can give your AI real-time context from: APIs Documents Databases Websit…  ( 29 min )

If you've been curious about how to build your own AI agents that can talk to APIs, connect with tools like databases, or even follow documentation you're in the right place. Microsoft has created something called MCP, which stands for Model‑Context‑Protocol. And to help you learn it step by step, they’ve made an amazing MCP Resources Hub on GitHub. In this blog, I’ll Walk you through what MCP is, why it matters, and how to use this hub to get started, even if you're new to AI development. What is MCP (Model‑Context‑Protocol)? Think of MCP like a communication bridge between your AI model and the outside world. Normally, when we chat with AI (like ChatGPT), it only knows what’s in its training data. But with MCP, you can give your AI real-time context from: APIs Documents Databases Websit…

Choosing the right hosting plan for your WordPress site on Azure App Service can feel overwhelming—but it doesn’t have to be. Whether you're just exploring WordPress or launching a high-traffic production site, we’ve created four tailored hosting plans to help you get started quickly and confidently. Let’s walk through how to pick the right plan for your needs. Which Hosting Plan Should You Choose? We’ve simplified the decision-making process with a clear recommendation based on your use case: Use CaseRecommended Plan Hobby or exploratory siteFree or Basic Small production websiteStandard High-load production websitePremium   💡 Important: Only the Premium plan supports High Availability (HA). This is the only setting that cannot be changed after deployment. If HA is a requirement,…  ( 25 min )

Choosing the right hosting plan for your WordPress site on Azure App Service can feel overwhelming—but it doesn’t have to be. Whether you're just exploring WordPress or launching a high-traffic production site, we’ve created four tailored hosting plans to help you get started quickly and confidently. Let’s walk through how to pick the right plan for your needs. Which Hosting Plan Should You Choose? We’ve simplified the decision-making process with a clear recommendation based on your use case: Use CaseRecommended Plan Hobby or exploratory siteFree or Basic Small production websiteStandard High-load production websitePremium   💡 Important: Only the Premium plan supports High Availability (HA). This is the only setting that cannot be changed after deployment. If HA is a requirement,…

We’re excited to announce that you can now orchestrate Azure Databricks Jobs from your Microsoft Fabric data pipelines! Databrick Jobs allow you to schedule and orchestrate a task or multiple tasks in a workflow in your Databricks workspace. Since any operation in Databricks can be a task, this means you can now run anything in … Continue reading “Orchestrate your Databricks Jobs with Fabric Data pipelines”  ( 6 min )

Stop Translating Docs Manually! Automate Your Global Reach with Co-op Translator v0.8 Series   Is your team or open-source project drowning in the endless cycle of manually translating documentation? Every update to your source content triggers a wave of tedious, error-prone work across multiple languages, slowing down knowledge sharing and hindering your global impact.   This challenge became particularly clear within large-scale Microsoft educational projects like the "For Beginners" series, where manual translation simply couldn't keep pace. A scalable, automated solution was needed to ensure valuable technical knowledge reaches learners and developers worldwide, breaking down language barriers that limit participation and slow innovation.   Co-op Translator, a Microsoft Azure open-sour…  ( 29 min )

Stop Translating Docs Manually! Automate Your Global Reach with Co-op Translator v0.8 Series   Is your team or open-source project drowning in the endless cycle of manually translating documentation? Every update to your source content triggers a wave of tedious, error-prone work across multiple languages, slowing down knowledge sharing and hindering your global impact.   This challenge became particularly clear within large-scale Microsoft educational projects like the "For Beginners" series, where manual translation simply couldn't keep pace. A scalable, automated solution was needed to ensure valuable technical knowledge reaches learners and developers worldwide, breaking down language barriers that limit participation and slow innovation.   Co-op Translator, a Microsoft Azure open-sour…

Introduction Azure AI Document Intelligence simplifies document data extraction, with features like query fields enabling targeted data retrieval. However, using these features with the C# SDK can be tricky. This guide highlights a real-world issue, provides a corrected implementation, and shares best practices for efficient usage.   Use case scenario During the cause of Azure AI Document Intelligence software engineering code tasks or review, many developers encountered an error while trying to extract fields like "FullName," "CompanyName," and "JobTitle" using `AnalyzeDocumentAsync`: The error might be similar to Inner Error: The parameter urlSource or base64Source is required. This is a challenge referred to as parameter errors and SDK changes. Most problematic code are looks like below…  ( 23 min )

Introduction Azure AI Document Intelligence simplifies document data extraction, with features like query fields enabling targeted data retrieval. However, using these features with the C# SDK can be tricky. This guide highlights a real-world issue, provides a corrected implementation, and shares best practices for efficient usage.   Use case scenario During the cause of Azure AI Document Intelligence software engineering code tasks or review, many developers encountered an error while trying to extract fields like "FullName," "CompanyName," and "JobTitle" using `AnalyzeDocumentAsync`: The error might be similar to Inner Error: The parameter urlSource or base64Source is required. This is a challenge referred to as parameter errors and SDK changes. Most problematic code are looks like below…

Introduction Azure AI Document Intelligence simplifies document data extraction, with features like query fields enabling targeted data retrieval. However, using these features with the C# SDK can be tricky. This guide highlights a real-world issue, provides a corrected implementation, and shares best practices for efficient usage.   Use case scenario During the cause of Azure AI Document Intelligence software engineering code tasks or review, many developers encountered an error while trying to extract fields like "FullName," "CompanyName," and "JobTitle" using `AnalyzeDocumentAsync`: The error might be similar to Inner Error: The parameter urlSource or base64Source is required. This is a challenge referred to as parameter errors and SDK changes. Most problematic code are looks like below…  ( 23 min )

Introduction Azure AI Document Intelligence simplifies document data extraction, with features like query fields enabling targeted data retrieval. However, using these features with the C# SDK can be tricky. This guide highlights a real-world issue, provides a corrected implementation, and shares best practices for efficient usage.   Use case scenario During the cause of Azure AI Document Intelligence software engineering code tasks or review, many developers encountered an error while trying to extract fields like "FullName," "CompanyName," and "JobTitle" using `AnalyzeDocumentAsync`: The error might be similar to Inner Error: The parameter urlSource or base64Source is required. This is a challenge referred to as parameter errors and SDK changes. Most problematic code are looks like below…

Table of Contents Abstract Introduction EDA Cloud-Compute scenarios Scenario 1: Burst to Azure from on-premises Data Center Scenario 2: “24x7 Single Set Workload” Scenario 3: "Data Center Supplement" Summary Abstract Azure NetApp Files (ANF) is transforming Electronic Design Automation (EDA) workflows in the cloud by delivering unparalleled performance, scalability, and efficiency. This blog explores how ANF addresses critical challenges in three cloud compute scenarios: Cloud Bursting, 24x7 All-in-Cloud, and Cloud-based Data Center Supplement. These solutions are tailored to optimize EDA processes, which rely on high-performance NFS file systems to design advanced semiconductor products. With the ability to support clusters exceeding 50,000 cores, ANF enhances productivity, shortens desi…  ( 45 min )

Table of Contents Abstract Introduction EDA Cloud-Compute scenarios Scenario 1: Burst to Azure from on-premises Data Center Scenario 2: “24x7 Single Set Workload” Scenario 3: "Data Center Supplement" Summary Abstract Azure NetApp Files (ANF) is transforming Electronic Design Automation (EDA) workflows in the cloud by delivering unparalleled performance, scalability, and efficiency. This blog explores how ANF addresses critical challenges in three cloud compute scenarios: Cloud Bursting, 24x7 All-in-Cloud, and Cloud-based Data Center Supplement. These solutions are tailored to optimize EDA processes, which rely on high-performance NFS file systems to design advanced semiconductor products. With the ability to support clusters exceeding 50,000 cores, ANF enhances productivity, shortens desi…

In today’s data-driven landscape, the ability to access and interpret information in a human-readable format is increasingly valuable. Being able to interact with and query your database in natural language is the game changer. In this post, we’ll walk through how to build a SQL agent using the Semantic Kernel framework to interact with a PostgreSQL database containing DVD rental data. I’ll explain how to define Semantic Kernel functions through plugins and how to incorporate them into agents. We’ll also look at how to set up these agents and guide them with well-structured instructions. Our example uses a PostgreSQL database that stores detailed information about DVD rentals. This sample database contains 15 tables and can be found here: PostgreSQL Sample Database. With a natural language…  ( 41 min )

In today’s data-driven landscape, the ability to access and interpret information in a human-readable format is increasingly valuable. Being able to interact with and query your database in natural language is the game changer. In this post, we’ll walk through how to build a SQL agent using the Semantic Kernel framework to interact with a PostgreSQL database containing DVD rental data. I’ll explain how to define Semantic Kernel functions through plugins and how to incorporate them into agents. We’ll also look at how to set up these agents and guide them with well-structured instructions. Our example uses a PostgreSQL database that stores detailed information about DVD rentals. This sample database contains 15 tables and can be found here: PostgreSQL Sample Database. With a natural language…

The architecture uses AKS to implement compute clusters of the applications that process high-volume batches of transactions. The applications receive the transactions in messages from Service Bus topics or queues. The topics and queues can be at Azure datacenters in different geographic regions, and multiple AKS clusters can read input from them.   Architecture       Workflow The numbered circles in the diagram correspond to the numbered steps in the following list. The architecture uses Service Bus topics and queues to organize the batch processing input and to pass it downstream for processing. Azure Load Balancer, a Layer 4 (TCP, UDP) load balancer, distributes incoming traffic among healthy instances of services defined in a load-balanced set. Load balancing and management of connec…  ( 59 min )

The architecture uses AKS to implement compute clusters of the applications that process high-volume batches of transactions. The applications receive the transactions in messages from Service Bus topics or queues. The topics and queues can be at Azure datacenters in different geographic regions, and multiple AKS clusters can read input from them.   Architecture       Workflow The numbered circles in the diagram correspond to the numbered steps in the following list. The architecture uses Service Bus topics and queues to organize the batch processing input and to pass it downstream for processing. Azure Load Balancer, a Layer 4 (TCP, UDP) load balancer, distributes incoming traffic among healthy instances of services defined in a load-balanced set. Load balancing and management of connec…

Enterprises today face several challenges in processing and extracting insights from multimodal data, like managing diverse data formats, ensuring data quality, and streamlining workflows efficiently. Ensuring the accuracy and usability of extracted insights often requires advanced AI techniques, while inefficiencies in managing large data volumes increase costs and delay results. Azure AI Content Understanding addresses these pain points by offering a unified solution to transform unstructured data into actionable insights, improve data accuracy with schema extraction and confidence scoring, and integrate seamlessly with Azure’s ecosystem to enhance efficiency and reduce costs. Content Understanding makes it easy to extract custom task-specific output without advanced GenAI skills. It ena…  ( 25 min )

Enterprises today face several challenges in processing and extracting insights from multimodal data, like managing diverse data formats, ensuring data quality, and streamlining workflows efficiently. Ensuring the accuracy and usability of extracted insights often requires advanced AI techniques, while inefficiencies in managing large data volumes increase costs and delay results. Azure AI Content Understanding addresses these pain points by offering a unified solution to transform unstructured data into actionable insights, improve data accuracy with schema extraction and confidence scoring, and integrate seamlessly with Azure’s ecosystem to enhance efficiency and reduce costs. Content Understanding makes it easy to extract custom task-specific output without advanced GenAI skills. It ena…

GraphQL has revolutionized the way developers interact with APIs by offering a more flexible and efficient alternative to REST. Before getting started , Create an API for GraphQL in Fabric and add data to use GraphQL in Fabric. At the heart of GraphQL are two core operations: queries and mutations. While they may look similar on the surface, they serve very different purposes. Let’s explain it in detail.  ( 7 min )

We’re excited to announce that native support for evaluating Data Agents through the Fabric SDK is now available in Preview. You can now run structured evaluations of your agent’s responses using Python — directly from notebooks or your own automation pipelines. Whether you’re validating accuracy before deploying to production, tuning prompts for better performance, or … Continue reading “Evaluate your Fabric Data Agents programmatically with the Python SDK (Preview)”  ( 7 min )

Introduction Have you ever thought about to build a Visual Studio Code extension as your capstone project? That’s what I did: Part 1 - Develop a VS Code Extension for Your Capstone Project. I have created a Visual Studio Code Extension, API Guardian, that identifies API endpoints in a project and checks their functionality before deployment. This solution was developed to help developers save time spent fixing issues caused by breaking or non-breaking changes and to alleviate the difficulties in performing maintenance due to unclear or outdated documentation. Let's build your very own extension! Now, let’s do it step by step.   Step 1 – Install the NPM package for generator-code Ensure Node.js is installed before proceeding. Verify by running node -v, which will display the installed versi…  ( 27 min )

Introduction Have you ever thought about to build a Visual Studio Code extension as your capstone project? That’s what I did: Part 1 - Develop a VS Code Extension for Your Capstone Project. I have created a Visual Studio Code Extension, API Guardian, that identifies API endpoints in a project and checks their functionality before deployment. This solution was developed to help developers save time spent fixing issues caused by breaking or non-breaking changes and to alleviate the difficulties in performing maintenance due to unclear or outdated documentation. Let's build your very own extension! Now, let’s do it step by step.   Step 1 – Install the NPM package for generator-code Ensure Node.js is installed before proceeding. Verify by running node -v, which will display the installed versi…

Introduction: The Era of AI Agents in JavaScript During the AI Agents Hackathon, one of the most anticipated sessions was presented by Wassim Chegham, Senior AI Developer Advocate for JavaScript at Microsoft. The topic? "How to Create Your First AI Agent with JavaScript and Azure AI Agent Service" — a powerful tool designed for modern developers looking to build AI-first applications with security, scalability, and productivity in mind. In this article, we explore the main highlights of the session, focusing on how you can create your own AI agent using JavaScript and Azure AI Agent Service. The video’s goal is clear: walk through the step-by-step process of creating AI agents using JavaScript and TypeScript with Azure AI Foundry, and explain all the key concepts behind this new developmen…  ( 35 min )

Introduction: The Era of AI Agents in JavaScript During the AI Agents Hackathon, one of the most anticipated sessions was presented by Wassim Chegham, Senior AI Developer Advocate for JavaScript at Microsoft. The topic? "How to Create Your First AI Agent with JavaScript and Azure AI Agent Service" — a powerful tool designed for modern developers looking to build AI-first applications with security, scalability, and productivity in mind. In this article, we explore the main highlights of the session, focusing on how you can create your own AI agent using JavaScript and Azure AI Agent Service. The video’s goal is clear: walk through the step-by-step process of creating AI agents using JavaScript and TypeScript with Azure AI Foundry, and explain all the key concepts behind this new developmen…

Shiva Shastri Sr Product Marketing Manager, Azure Migrate—Product & Ecosystem. Cost-effective and sustainable innovation. In today's rapidly evolving digital landscape, businesses are constantly seeking ways to stay competitive through innovations while managing costs. By leveraging the power of the cloud, organizations can achieve cost-effectiveness and foster sustainable innovation. By transitioning to Azure, any organization can achieve greater financial flexibility, operational efficiency, and gain access to innovations that provide a competitive edge in the marketplace. Collocating application resources and data is essential for optimal performance and return on investment (ROI). Once in Azure, secure and responsible AI can help you with insights and actions that lead to better outcom…  ( 27 min )

Shiva Shastri Sr Product Marketing Manager, Azure Migrate—Product & Ecosystem. Cost-effective and sustainable innovation. In today's rapidly evolving digital landscape, businesses are constantly seeking ways to stay competitive through innovations while managing costs. By leveraging the power of the cloud, organizations can achieve cost-effectiveness and foster sustainable innovation. By transitioning to Azure, any organization can achieve greater financial flexibility, operational efficiency, and gain access to innovations that provide a competitive edge in the marketplace. Collocating application resources and data is essential for optimal performance and return on investment (ROI). Once in Azure, secure and responsible AI can help you with insights and actions that lead to better outcom…

The strategic partnership between Microsoft Azure and Red Hat continues to flourish in 2025, with both companies showcasing their joint innovations at two major tech events: Microsoft Build (May 19-22 in Seattle) and Red Hat Summit 2025. This collaboration represents one of tech's most impactful partnerships, combining Microsoft's cloud expertise with Red Hat's open-source leadership to create solutions that drive digital transformation across industries. Microsoft Build 2025: AI Innovation Meets Open Source Microsoft Build 2025 will take place from May 19-22 at the Seattle Convention Center, bringing together developers, creators, and AI innovators from around the world. This year's event has been extended to four days, offering more opportunities for learning and networking. A key highl…  ( 34 min )

The strategic partnership between Microsoft Azure and Red Hat continues to flourish in 2025, with both companies showcasing their joint innovations at two major tech events: Microsoft Build (May 19-22 in Seattle) and Red Hat Summit 2025. This collaboration represents one of tech's most impactful partnerships, combining Microsoft's cloud expertise with Red Hat's open-source leadership to create solutions that drive digital transformation across industries. Microsoft Build 2025: AI Innovation Meets Open Source Microsoft Build 2025 will take place from May 19-22 at the Seattle Convention Center, bringing together developers, creators, and AI innovators from around the world. This year's event has been extended to four days, offering more opportunities for learning and networking. A key highl…

Software development is changing fast, and Java developers are right in the middle of it – especially when it comes to using Artificial Intelligence (AI) in their apps. This report brings together feedback from 647 Java professionals to show where things stand and what is possible as Java and AI come together. One of the […] The post The State of Coding the Future with Java and AI – May 2025 appeared first on Microsoft for Java Developers.  ( 44 min )

Additional contributors: Venkat Ramakrishnan, Amir Jafari, Wilson Lee, Maraki Ketema As organizations accelerate their hybrid cloud adoption strategies, Oracle Database@Azure has emerged as a critical platform for running Oracle database workloads using Exadata, Autonomous, Exadata Exascale and Base databases for the enterprise. However, deriving real-time, AI-powered insights in hybrid settings has long been challenging due … Continue reading “Accelerate AI on Oracle Databases with Open Mirroring, Fabric Data Agent, and Azure AI Foundry”  ( 9 min )

We’re excited to introduce Copilot for SQL Analytics Endpoint, now in preview – a transformative, AI-powered assistant built to change how you query, explore, and analyze data within Microsoft Fabric’s SQL experience. With Copilot integrated directly into the SQL Analytics Endpoint, users can now express intent in natural language and instantly receive ready-to-run T-SQL. Whether … Continue reading “Announcing Copilot for SQL Analytics Endpoint in Microsoft Fabric (Preview)”  ( 7 min )

Build your first Agent App Agentic AI is changing how we build intelligent apps - enabling software to reason, plan, and act for us. Learning to build AI agents is quickly becoming a must-have skill for anyone working with AI. Self-Guided Workshop Try our self-guided “Build your code-first agent with Azure AI Foundry” workshop to get hands-on with Azure AI Agent Service. You’ll learn to build, deploy, and interact with agents using Azure’s powerful tools. What is Azure AI Agent Service? Azure AI Agent Service lets you create, orchestrate, and manage AI-powered agents that can handle complex tasks, integrate with tools, and deploy securely. What Will You Learn? The basics of agentic AI apps and how they differ from traditional apps How to set up your Azure environment How to build your first agent How to test and interact with your agent Advanced features like tool integration and memory management Who Is This For? Anyone interested in building intelligent, goal-oriented agents — developers, data scientists, students, and AI enthusiasts. No prior experience with Azure AI Agent Service required. How Does the Workshop Work? Tip: Select the self-guided tab in Getting Started for the right instructions. Step-by-step guides at your own pace Code samples and templates Real-world scenarios Get Started See what agentic AI can do for you with the self-guided “Build your code-first agent with Azure AI Foundry” workshop. Build practical skills in one of AI’s most exciting areas. Try the workshop and start building agents that make a difference! Additional Resources Azure AI Foundry Documentation Azure AI Agent Service Overview Questions or feedback Questions or feedback? Visit the issues page. Happy learning and building with Azure AI Agent Service!  ( 21 min )

Build your first Agent App Agentic AI is changing how we build intelligent apps - enabling software to reason, plan, and act for us. Learning to build AI agents is quickly becoming a must-have skill for anyone working with AI. Self-Guided Workshop Try our self-guided “Build your code-first agent with Azure AI Foundry” workshop to get hands-on with Azure AI Agent Service. You’ll learn to build, deploy, and interact with agents using Azure’s powerful tools. What is Azure AI Agent Service? Azure AI Agent Service lets you create, orchestrate, and manage AI-powered agents that can handle complex tasks, integrate with tools, and deploy securely. What Will You Learn? The basics of agentic AI apps and how they differ from traditional apps How to set up your Azure environment How to build your first agent How to test and interact with your agent Advanced features like tool integration and memory management Who Is This For? Anyone interested in building intelligent, goal-oriented agents — developers, data scientists, students, and AI enthusiasts. No prior experience with Azure AI Agent Service required. How Does the Workshop Work? Tip: Select the self-guided tab in Getting Started for the right instructions. Step-by-step guides at your own pace Code samples and templates Real-world scenarios Get Started See what agentic AI can do for you with the self-guided “Build your code-first agent with Azure AI Foundry” workshop. Build practical skills in one of AI’s most exciting areas. Try the workshop and start building agents that make a difference! Additional Resources Azure AI Foundry Documentation Azure AI Agent Service Overview Questions or feedback Questions or feedback? Visit the issues page. Happy learning and building with Azure AI Agent Service!

At the time of this writing, Azure AI Agent Service does not offer turnkey integration with Model Context Protocol (MCP) Servers. Discussed here is a solution that helps to leverage MCP's powerful capabilities while working within the Azure ecosystem. The integration approach piggybacks on the Function integration capability in the Azure AI Agent Service. By utilizing an MCP Client to discover and register tools from an MCP Server as Functions with the Agent Service, we create a seamless integration layer between the two systems. Built using the Microsoft Bot Framework, this application can be published as an AI Assistant across numerous channels like Microsoft Teams, Slack, and others. For development and testing purposes, we've used the Bot Framework Emulator to run and validate the appl…  ( 29 min )

At the time of this writing, Azure AI Agent Service does not offer turnkey integration with Model Context Protocol (MCP) Servers. Discussed here is a solution that helps to leverage MCP's powerful capabilities while working within the Azure ecosystem. The integration approach piggybacks on the Function integration capability in the Azure AI Agent Service. By utilizing an MCP Client to discover and register tools from an MCP Server as Functions with the Agent Service, we create a seamless integration layer between the two systems. Built using the Microsoft Bot Framework, this application can be published as an AI Assistant across numerous channels like Microsoft Teams, Slack, and others. For development and testing purposes, we've used the Bot Framework Emulator to run and validate the appl…

ASC Technologies, a leader in compliance recording and AI-driven data analytics, provides cutting-edge software solutions for capturing and analyzing communication channels. Their innovative technology empowers more than 500 customers worldwide to record communications legally while extracting valuable insights and helping to prevent fraudulent activities. Many of their customers operate in heavily regulated industries where compliance recording is mandatory. These organizations rely on the ability to consolidate and analyze information shared across multiple channels including voice recordings, chat logs, speaker recognition, video analysis, and document and screen activity. As ASC’s customer base expanded, and their clients accumulated millions of calls and vast amounts of conversation m…  ( 25 min )

ASC Technologies, a leader in compliance recording and AI-driven data analytics, provides cutting-edge software solutions for capturing and analyzing communication channels. Their innovative technology empowers more than 500 customers worldwide to record communications legally while extracting valuable insights and helping to prevent fraudulent activities. Many of their customers operate in heavily regulated industries where compliance recording is mandatory. These organizations rely on the ability to consolidate and analyze information shared across multiple channels including voice recordings, chat logs, speaker recognition, video analysis, and document and screen activity. As ASC’s customer base expanded, and their clients accumulated millions of calls and vast amounts of conversation m…

In today’s data-driven world, organizations are increasingly turning to AI for document understanding. Whether it's extracting invoices, contracts, ID cards, or complex forms, Azure Document Intelligence (formerly known as Form Recognizer) provides a robust, AI-powered solution for automated document processing. But what happens when you want to scale, secure, and load balance your document intelligence backend for high availability and enterprise-grade integration? Enter Azure API Management (APIM) — your gateway to efficient, scalable API orchestration. In this blog, we’ll explore how to integrate Azure Document Intelligence with APIM using a load-balanced architecture that works seamlessly with the Document Intelligence SDK — without rewriting your application logic. Azure Doc Intellige…  ( 36 min )

In today’s data-driven world, organizations are increasingly turning to AI for document understanding. Whether it's extracting invoices, contracts, ID cards, or complex forms, Azure Document Intelligence (formerly known as Form Recognizer) provides a robust, AI-powered solution for automated document processing. But what happens when you want to scale, secure, and load balance your document intelligence backend for high availability and enterprise-grade integration? Enter Azure API Management (APIM) — your gateway to efficient, scalable API orchestration. In this blog, we’ll explore how to integrate Azure Document Intelligence with APIM using a load-balanced architecture that works seamlessly with the Document Intelligence SDK — without rewriting your application logic. Azure Doc Intellige…

This post will cover how to remotely connect to the JVM when running on Azure App Service with Java.  ( 3 min )

Introduction Azure Container Apps offers a robust platform for deploying microservices and containerized applications. When integrating with Azure Application Gateway, an internal container app environment can be accessed via public internet. Users often bind custom domains to enhance accessibility and user experience. A common challenge arises when we bind the custom domain on Application Gateway and try to access container app. Container app is acting as a middleware service and needs to forward request to another API server or finish authentication process, users may encountered HTTP 403 forbidden error which is caused by hostname/redirect URL mismatch. What's more, you definitely don't want to expose your backend service default domain. This blog explores these challenges and offers pr…  ( 22 min )

Introduction Azure Container Apps offers a robust platform for deploying microservices and containerized applications. When integrating with Azure Application Gateway, an internal container app environment can be accessed via public internet. Users often bind custom domains to enhance accessibility and user experience. A common challenge arises when we bind the custom domain on Application Gateway and try to access container app. Container app is acting as a middleware service and needs to forward request to another API server or finish authentication process, users may encountered HTTP 403 forbidden error which is caused by hostname/redirect URL mismatch. What's more, you definitely don't want to expose your backend service default domain. This blog explores these challenges and offers pr…

The May 2025 release includes updates in the Python Environments extension, a new color picker added by Pylance, branch coverage support, and more! The post Python in Visual Studio Code – May 2025 Release appeared first on Microsoft for Python Developers Blog.  ( 24 min )

We are less than 18 months away from the retirement of Exchange Web Services. Start planning your migration from EWS to Microsoft Graph. The post Exchange Web Services code analyzer and usage report appeared first on Microsoft 365 Developer Blog.  ( 24 min )

With this practical guide, you now know how to secure your Kubernetes cluster using the structured-authentication feature, offering flexible integration with any JWT-compliant token provider. The post JWT it like it’s hot: A practical guide for Kubernetes Structured Authentication appeared first on Microsoft Open Source Blog.  ( 16 min )

Shortcut cache and on-prem gateway support are now generally available (GA) Shortcut cache Shortcuts in OneLake allow you to quickly and easily source data from external cloud providers and use it across all Fabric workloads such as Power BI reports, SQL, Spark and Kusto.  However, each time these workloads read data from cross-cloud sources, the … Continue reading “Shortcut cache and on-prem gateway support (Generally Available)”  ( 6 min )

Shortcuts in OneLake provide a quick and easy way to make your data available in Fabric without having to copy it.  Simply create a new shortcut and your data is instantly available to all Fabric workloads. When you first create a new shortcut, you also set up a shared cloud connection. These are the same connections … Continue reading “Manage connections for shortcuts”  ( 6 min )

I’ve made this Simple FastAPI AZD template for Azure App Service to help you get to the fun part, and to cut out all the extra infrastructure that you don’t necessarily want or need. This FastAPI template for Azure App Service gives you all the infrastructure as code to deploy a basic “Hello World” FastAPI web app that you can spin up using AZD (Azure Developer SDK) with just three commands. How to do it To get started, you just need to install AZD, a command-line tool you can use right there in VS Code. Then you’re ready to grab the template and deploy. Run these commands and follow the prompts as you go.  Grab our new simple FastAPI template for Azure App Service: azd init --template Azure-Samples/azd-simple-fastapi-appservice It will ask you to give your environment a name. This will b…  ( 33 min )

I’ve made this Simple FastAPI AZD template for Azure App Service to help you get to the fun part, and to cut out all the extra infrastructure that you don’t necessarily want or need. This FastAPI template for Azure App Service gives you all the infrastructure as code to deploy a basic “Hello World” FastAPI web app that you can spin up using AZD (Azure Developer SDK) with just three commands. How to do it To get started, you just need to install AZD, a command-line tool you can use right there in VS Code. Then you’re ready to grab the template and deploy. Run these commands and follow the prompts as you go.  Grab our new simple FastAPI template for Azure App Service: azd init --template Azure-Samples/azd-simple-fastapi-appservice It will ask you to give your environment a name. This will b…

AI distillation is a powerful technique in machine learning that involves transferring knowledge from a large, complex model (often called the "teacher") to a smaller, more efficient "student" model. The goal is to retain the performance and accuracy of the larger model while drastically reducing computational requirements, making AI systems faster, cheaper, and more deployable especially in real-time or resource-constrained environments. In this post, we'll explore what AI distillation is, why it's gaining traction, and how it's being used to bring the power of advanced AI models to everyday applications. Stored completions in Azure OpenAI's AI Foundry provide a structured way to capture and reuse high-quality model responses, streamlining the model distillation process. By logging curate…  ( 38 min )

AI distillation is a powerful technique in machine learning that involves transferring knowledge from a large, complex model (often called the "teacher") to a smaller, more efficient "student" model. The goal is to retain the performance and accuracy of the larger model while drastically reducing computational requirements, making AI systems faster, cheaper, and more deployable especially in real-time or resource-constrained environments. In this post, we'll explore what AI distillation is, why it's gaining traction, and how it's being used to bring the power of advanced AI models to everyday applications. Stored completions in Azure OpenAI's AI Foundry provide a structured way to capture and reuse high-quality model responses, streamlining the model distillation process. By logging curate…

Are you looking to build custom Copilots but unsure about the differences between Copilot Studio and Azure AI Foundry? As a Microsoft Technical Trainer with over a decade of experience, I've spent the last 18 months focusing on Azure AI Solutions and Copilot. Through numerous workshops, I've seen firsthand how customers benefit from AI solutions beyond Microsoft Copilot. Microsoft 365 Copilot Chat offers seamless integration with Generative AI for tasks like document creation, content summarization, and insights from M365 solutions such as Email, OneDrive, SharePoint, and Teams. It ensures compliance with organizational security, governance, and privacy policies, making it ideal for immediate AI assistance without customization. On the other hand, platforms like Copilot Studio and Azure AI…  ( 37 min )

Are you looking to build custom Copilots but unsure about the differences between Copilot Studio and Azure AI Foundry? As a Microsoft Technical Trainer with over a decade of experience, I've spent the last 18 months focusing on Azure AI Solutions and Copilot. Through numerous workshops, I've seen firsthand how customers benefit from AI solutions beyond Microsoft Copilot. Microsoft 365 Copilot Chat offers seamless integration with Generative AI for tasks like document creation, content summarization, and insights from M365 solutions such as Email, OneDrive, SharePoint, and Teams. It ensures compliance with organizational security, governance, and privacy policies, making it ideal for immediate AI assistance without customization. On the other hand, platforms like Copilot Studio and Azure AI…

New Models Available!   We’re excited to announce the preview availability of the following Azure OpenAI Service models for use in the Azure AI Agent Service, starting 5/7:  o1  o3-mini  gpt-4.1  gpt-4.1-mini  gpt-4.1-nano    Azure OpenAI o-Series Models  Azure OpenAI o-series models are designed to tackle reasoning and problem-solving tasks with increased focus and capability. These models spend more time processing and understanding the user's request, making them exceptionally strong in areas like science, coding, and math compared to previous iterations.    o1: The most capable model in the o1 series, offering enhanced reasoning abilities.  o3 (coming soon): The most capable reasoning model in the o model series, and the first one to offer full tools support for agentic so…  ( 30 min )

New Models Available!   We’re excited to announce the preview availability of the following Azure OpenAI Service models for use in the Azure AI Agent Service, starting 5/7:  o1  o3-mini  gpt-4.1  gpt-4.1-mini  gpt-4.1-nano    Azure OpenAI o-Series Models  Azure OpenAI o-series models are designed to tackle reasoning and problem-solving tasks with increased focus and capability. These models spend more time processing and understanding the user's request, making them exceptionally strong in areas like science, coding, and math compared to previous iterations.    o1: The most capable model in the o1 series, offering enhanced reasoning abilities.  o3 (coming soon): The most capable reasoning model in the o model series, and the first one to offer full tools support for agentic so…

Starting June 28, 2025, the European Accessibility Act (EAA) takes effect, requiring all digital products and services offered to EU customers to meet comprehensive accessibility standards. If you're developing Office Add-ins, this may impact you. This blog explains what you need to know to ensure your Office Add-ins are compliant. The post Prepare your Office Add-in for the European Accessibility Act (EAA) appeared first on Microsoft 365 Developer Blog.  ( 23 min )

Software development is changing fast, and Java developers are right in the middle of it - especially when it comes to using Artificial Intelligence (AI) in their apps. This report brings together feedback from 647 Java professionals to show where things stand and what is possible as Java and AI come together. One of the biggest takeaways is this: Java developers do not need to be experts in AI, machine learning, or Python. With tools like the Model Context Protocol (MCP) Java SDK, Spring AI, and LangChain4j, they can start adding smart features to their apps using the skills they already have. Whether it is making recommendations, spotting fraud, supporting natural language search or a world of possibilities, AI can be part of everyday Java development. The report walks through real-world…  ( 102 min )

Software development is changing fast, and Java developers are right in the middle of it - especially when it comes to using Artificial Intelligence (AI) in their apps. This report brings together feedback from 647 Java professionals to show where things stand and what is possible as Java and AI come together. One of the biggest takeaways is this: Java developers do not need to be experts in AI, machine learning, or Python. With tools like the Model Context Protocol (MCP) Java SDK, Spring AI, and LangChain4j, they can start adding smart features to their apps using the skills they already have. Whether it is making recommendations, spotting fraud, supporting natural language search or a world of possibilities, AI can be part of everyday Java development. The report walks through real-world…

Starting on June 9, 2025, all Power BI and Fabric capacity SKUs will support XMLA read/write operations by default. This change is intended to assist customers using XMLA-based tools to create, edit, and maintain semantic models, such as DAX Query View in the web, Live Editing in Power BI Desktop, SQL Server Management Studio (SSMS), … Continue reading “Enabling broader adoption of XMLA-based tools and scenarios”  ( 5 min )

API Guardian - My Capstone Project As software and APIs evolve, developers encounter significant difficulties in maintaining and updating API endpoints. Breaking changes can lead to system instability, while outdated or unclear documentation makes maintenance less efficient. These challenges are further compounded by the time-consuming nature of updating dependencies and the tendency to prioritize new features over maintenance tasks. The absence of effective tools and processes to tackle these issues reduces overall productivity and developer efficiency. To address this, API Guardian was created as a Visual Studio Code extension that identifies API endpoints in a project and checks their functionality before deployment. This solution was developed to help developers save time spent fixing …  ( 26 min )

API Guardian - My Capstone Project As software and APIs evolve, developers encounter significant difficulties in maintaining and updating API endpoints. Breaking changes can lead to system instability, while outdated or unclear documentation makes maintenance less efficient. These challenges are further compounded by the time-consuming nature of updating dependencies and the tendency to prioritize new features over maintenance tasks. The absence of effective tools and processes to tackle these issues reduces overall productivity and developer efficiency. To address this, API Guardian was created as a Visual Studio Code extension that identifies API endpoints in a project and checks their functionality before deployment. This solution was developed to help developers save time spent fixing …

Scenario: In case of encountering error in API management, request tracing is an invaluable feature that serves as a debugger. It allows for tracking the flow of requests as they pass through various policy logic, providing detailed insights into the complete API Management (APIM) processing. Here is a link if you would like to read more on how to enable request tracing in API management.  However, it is the most common way to debug your API, let's assume a real-life scenario where you encounter a sporadic error or unexpected response while processing the live APIM calls and need to drill down the issue. In such cases, attaching a debugger or running request traces can be challenging, especially when the issue is intermittent or requires checking the specific code logic. This often necessi…  ( 29 min )

Scenario: In case of encountering error in API management, request tracing is an invaluable feature that serves as a debugger. It allows for tracking the flow of requests as they pass through various policy logic, providing detailed insights into the complete API Management (APIM) processing. Here is a link if you would like to read more on how to enable request tracing in API management.  However, it is the most common way to debug your API, let's assume a real-life scenario where you encounter a sporadic error or unexpected response while processing the live APIM calls and need to drill down the issue. In such cases, attaching a debugger or running request traces can be challenging, especially when the issue is intermittent or requires checking the specific code logic. This often necessi…

Get started with Nested App Authentication, a modern protocol for simplifying authentication for Personal Tab Teams apps that run across Microsoft 365. The post Nested App Authentication: Now generally available across Microsoft 365 appeared first on Microsoft 365 Developer Blog.  ( 23 min )

We’re excited to announce the release candidate of the Semantic Kernel for Java Agents API! This marks a major step forward in bringing the power of intelligent agents to Java developers, enabling them to build rich, contextual, and interactive AI experiences using the Semantic Kernel framework. What Are Agents in Semantic Kernel? Agents are intelligent, autonomous […] The post RC1: Semantic Kernel for Java Agents API appeared first on Semantic Kernel.  ( 22 min )

In today's data-driven business environment, audit teams often spend weeks poring over logs and databases to verify spending and billing information. This time-consuming process is ripe for automation. But is there a way to implement AI solutions without getting lost in complex technical frameworks? While tools like LangChain, Semantic Kernel, and AutoGen offer powerful AI agent capabilities, sometimes you need a straightforward solution that just works.  So, what's the answer for teams seeking simplicity without sacrificing effectiveness? This tutorial will show you how to use Azure AI Agent Service to build an AI agent that can directly access your Postgres database to streamline audit workflows. No complex chains or graphs required, just a practical solution to get your audit process au…  ( 43 min )

In today's data-driven business environment, audit teams often spend weeks poring over logs and databases to verify spending and billing information. This time-consuming process is ripe for automation. But is there a way to implement AI solutions without getting lost in complex technical frameworks? While tools like LangChain, Semantic Kernel, and AutoGen offer powerful AI agent capabilities, sometimes you need a straightforward solution that just works.  So, what's the answer for teams seeking simplicity without sacrificing effectiveness? This tutorial will show you how to use Azure AI Agent Service to build an AI agent that can directly access your Postgres database to streamline audit workflows. No complex chains or graphs required, just a practical solution to get your audit process au…

We are pleased to announce that hubs and workspaces is now generally available on Azure machine learning allowing users to use hub for team’s collaboration environment for machine learning applications.  Azure Hubs and Workspaces provide a centralized platform capability for Azure Machine Learning. This feature enables developers to innovate faster by creating project workspaces and accessing shared company resources without needing repeated assistance from IT administrators.  Quick Model Building and Experimentation without IT bottleneck  Hubs and Workspaces in Azure Machine Learning provide a centralized solution for managing machine learning resources. Hubs act as a central resource management construct that oversees security, connectivity, computing resources, and team quotas. Once cre…  ( 26 min )

We are pleased to announce that hubs and workspaces is now generally available on Azure machine learning allowing users to use hub for team’s collaboration environment for machine learning applications.  Azure Hubs and Workspaces provide a centralized platform capability for Azure Machine Learning. This feature enables developers to innovate faster by creating project workspaces and accessing shared company resources without needing repeated assistance from IT administrators.  Quick Model Building and Experimentation without IT bottleneck  Hubs and Workspaces in Azure Machine Learning provide a centralized solution for managing machine learning resources. Hubs act as a central resource management construct that oversees security, connectivity, computing resources, and team quotas. Once cre…

Discover the new and improved Teams AI Library, designed to help developers create even more powerful agents for Microsoft Teams. The post Announcing the updated Teams AI Library and MCP support appeared first on Microsoft 365 Developer Blog.  ( 24 min )

We’re launching a quick survey to gather your feedback on Azure Monitor Log Analytics. Your input directly impacts our product roadmap and helps us prioritize the features and improvements that matter most to you. The survey takes just a few minutes, and your responses will remain confidential. Take the Survey   New to Log Analytics? Start here: Get Started with Azure Monitor Log Analytics Overview of Log Analytics in Azure Monitor - Azure Monitor | Microsoft Learn For questions or additional feedback, feel free to reach out to Noyablanga@microsoft.com.Thank you for being part of this journey!  ( 18 min )

We’re launching a quick survey to gather your feedback on Azure Monitor Log Analytics. Your input directly impacts our product roadmap and helps us prioritize the features and improvements that matter most to you. The survey takes just a few minutes, and your responses will remain confidential. Take the Survey   New to Log Analytics? Start here: Get Started with Azure Monitor Log Analytics Overview of Log Analytics in Azure Monitor - Azure Monitor | Microsoft Learn For questions or additional feedback, feel free to reach out to Noyablanga@microsoft.com.Thank you for being part of this journey!

This post refers to issues where you may see revisions “stuck in ‘Activating’ state” when using Azure Container Apps - and what are some common causes and explainations behind this.  ( 6 min )

We often hear about Large Language Models (LLMs) like GPT-4 having billions of parameters and being trained on massive datasets. But have you ever wondered: Where is all that data actually stored? And more fundamentally, how does a computer even store knowledge in the first place? Let’s take a journey from the world of electric charges to the vast neural networks powering today’s AI. Data at the Most Basic Level: 0s and 1s At its core, all digital data is just binary — a series of 0s and 1s. These bits are represented physically using electric charges or magnetic states: In RAM or CPU/GPU memory, bits are stored using transistors and capacitors that are either charged (1) or not charged (0). In SSDs, data is stored using floating-gate transistors that trap electrons to represent binary st…  ( 23 min )

We often hear about Large Language Models (LLMs) like GPT-4 having billions of parameters and being trained on massive datasets. But have you ever wondered: Where is all that data actually stored? And more fundamentally, how does a computer even store knowledge in the first place? Let’s take a journey from the world of electric charges to the vast neural networks powering today’s AI. Data at the Most Basic Level: 0s and 1s At its core, all digital data is just binary — a series of 0s and 1s. These bits are represented physically using electric charges or magnetic states: In RAM or CPU/GPU memory, bits are stored using transistors and capacitors that are either charged (1) or not charged (0). In SSDs, data is stored using floating-gate transistors that trap electrons to represent binary st…

With Fabric Events general availability, the role of Activator expands from setting notifications and acting on your data in real time to becoming an orchestration centerpiece. Activator acts as a connecting tissue enabling complex event-driven and data-driven flows. Let’s look at a very common architecture we often see our customers implement: In this architecture we … Continue reading “Activator as an Orchestrator of the Fabric Event Driven flows”  ( 6 min )

Task flows feature is now generally available! Task flows streamline the design of your data solutions and ensure consistency between design and development efforts. It also allows you to navigate items and manage your workspace more easily, even as it becomes more complex over time. Since its preview last May, we have received a great … Continue reading “Task flows in Microsoft Fabric (Generally Available)”  ( 6 min )

Hi everyone, Shivam Goyal here! This marks the final installment in our AI Agents for Beginners series, based on the awesome repository (link to the repo). I hope you've enjoyed this journey into the world of agentic AI! In previous posts ([links to parts 1-9 at the end]), we've covered the fundamentals and key design patterns. Now, let's explore the practical considerations of deploying AI agents to production, focusing on performance, cost management, and evaluation. As an active member of the AI community, I'm excited to share these insights to help you bring your agentic AI projects to life. From Lab to Production: Key Considerations Successfully deploying AI agents requires careful planning and attention to detail. We need to consider: How to plan the deployment of your AI Agent to p…  ( 26 min )

Hi everyone, Shivam Goyal here! This marks the final installment in our AI Agents for Beginners series, based on the awesome repository (link to the repo). I hope you've enjoyed this journey into the world of agentic AI! In previous posts ([links to parts 1-9 at the end]), we've covered the fundamentals and key design patterns. Now, let's explore the practical considerations of deploying AI agents to production, focusing on performance, cost management, and evaluation. As an active member of the AI community, I'm excited to share these insights to help you bring your agentic AI projects to life. From Lab to Production: Key Considerations Successfully deploying AI agents requires careful planning and attention to detail. We need to consider: How to plan the deployment of your AI Agent to p…

When integrating Azure resources like Logic Apps, Function Apps, or Azure VMs with SharePoint Online, you often need secure and granular access control. Rather than handling credentials manually, Managed Identity is the recommended approach to securely authenticate to Microsoft Graph and access SharePoint resources. High-level steps: Step 1: Enable Managed Identity (or App Registration) Step 2: Grant Sites.Selected Permission in Microsoft Entra ID Step 3: Assign SharePoint Site-Level Permission Step 1: Enable Managed Identity (or App Registration) For your Azure resource (e.g., Logic App): Navigate to the Azure portal. Go to the resource (e.g., Logic App). Under Identity, enable System-assigned Managed Identity. Note the Object ID and Client ID (you’ll need the Client ID later). Alternat…  ( 23 min )

When integrating Azure resources like Logic Apps, Function Apps, or Azure VMs with SharePoint Online, you often need secure and granular access control. Rather than handling credentials manually, Managed Identity is the recommended approach to securely authenticate to Microsoft Graph and access SharePoint resources. High-level steps: Step 1: Enable Managed Identity (or App Registration) Step 2: Grant Sites.Selected Permission in Microsoft Entra ID Step 3: Assign SharePoint Site-Level Permission Step 1: Enable Managed Identity (or App Registration) For your Azure resource (e.g., Logic App): Navigate to the Azure portal. Go to the resource (e.g., Logic App). Under Identity, enable System-assigned Managed Identity. Note the Object ID and Client ID (you’ll need the Client ID later). Alternat…

Today we’re excited to welcome Jarre Nejatyab as a guest blog to highlight a technical deep dive on orchestrating AI Agents with Semantic Kernel Plugins. In the rapidly evolving world of Large Language Models (LLMs), orchestrating specialized AI agents has become crucial for building sophisticated cognitive architectures capable of complex reasoning and task execution. While […] The post Guest Blog: Orchestrating AI Agents with Semantic Kernel Plugins: A Technical Deep Dive appeared first on Semantic Kernel.  ( 28 min )

Azure OpenAI has expanded its speech recognition capabilities with two powerful models: GPT-4o-transcribe and GPT-4o-mini-transcribe. These models also leverage WebSocket connections to enable real-time transcription of audio streams, providing developers with cutting-edge tools for speech-to-text applications. In this technical blog, we'll explore how these models work and demonstrate a practical implementation using Python. Understanding OpenAI's Realtime Transcription API Unlike the regular REST API for audio transcription, Azure OpenAI's Realtime API enables continuous streaming of audio data through WebSockets or WebRTC connections. This approach is particularly valuable for applications requiring immediate transcription feedback, such as live captioning, meeting transcription, or voi…  ( 31 min )

Azure OpenAI has expanded its speech recognition capabilities with two powerful models: GPT-4o-transcribe and GPT-4o-mini-transcribe. These models also leverage WebSocket connections to enable real-time transcription of audio streams, providing developers with cutting-edge tools for speech-to-text applications. In this technical blog, we'll explore how these models work and demonstrate a practical implementation using Python. Understanding OpenAI's Realtime Transcription API Unlike the regular REST API for audio transcription, Azure OpenAI's Realtime API enables continuous streaming of audio data through WebSockets or WebRTC connections. This approach is particularly valuable for applications requiring immediate transcription feedback, such as live captioning, meeting transcription, or voi…

We are excited to share that we have released a new Microsoft Fabric pipeline template that helps you easily send the results from Azure AI Content Understanding into a Fabric Lakehouse! This template makes it easier than ever to harvest information from multimodal content and turn it into structured data using Content Understanding and perform further analysis in Microsoft Fabric. Whether you are looking to extract insights from a contract, call transcript, or video footage, this template simplifies the process and gives you fast access to Fabric’s powerful data tools. Why It Matters Azure AI Content Understanding uses powerful large language models (LLMs) to extract key information from documents, videos, audio, and image files. For example, it can identify key phrases in documents, extract tables from invoices, or generate video chapters and summaries. This template lets you seamlessly feed structured JSON outputs from Content Understanding into a Fabric Lakehouse, where you can immediately use Power BI, Dataflows, and other tools to analyze and make sense of the data.   Key Benefits Quick Setup: Move from unstructured content to structured data in no time—no complicated setup required! Seamless Integration: Connect Azure AI and Microsoft Fabric effortlessly. Secure & Scalable: Every component is built on Microsoft’s cloud, ensuring your data is safe and scalable as your needs grow. Try It Now You can find the template and setup instructions on GitHub. We would love to hear how you are using it! Feel free to leave any questions or feedback in the comments below or send us an email.  Resources & Documentation Explore the following resources to learn more about Azure AI Content Understanding and Microsoft Fabric Azure Content Understanding Overview Microsoft Fabric Overview Azure Content Understanding in Azure AI Foundry Azure Content Understanding FAQs  ( 21 min )

We are excited to share that we have released a new Microsoft Fabric pipeline template that helps you easily send the results from Azure AI Content Understanding into a Fabric Lakehouse! This template makes it easier than ever to harvest information from multimodal content and turn it into structured data using Content Understanding and perform further analysis in Microsoft Fabric. Whether you are looking to extract insights from a contract, call transcript, or video footage, this template simplifies the process and gives you fast access to Fabric’s powerful data tools. Why It Matters Azure AI Content Understanding uses powerful large language models (LLMs) to extract key information from documents, videos, audio, and image files. For example, it can identify key phrases in documents, extract tables from invoices, or generate video chapters and summaries. This template lets you seamlessly feed structured JSON outputs from Content Understanding into a Fabric Lakehouse, where you can immediately use Power BI, Dataflows, and other tools to analyze and make sense of the data.   Key Benefits Quick Setup: Move from unstructured content to structured data in no time—no complicated setup required! Seamless Integration: Connect Azure AI and Microsoft Fabric effortlessly. Secure & Scalable: Every component is built on Microsoft’s cloud, ensuring your data is safe and scalable as your needs grow. Try It Now You can find the template and setup instructions on GitHub. We would love to hear how you are using it! Feel free to leave any questions or feedback in the comments below or send us an email.  Resources & Documentation Explore the following resources to learn more about Azure AI Content Understanding and Microsoft Fabric Azure Content Understanding Overview Microsoft Fabric Overview Azure Content Understanding in Azure AI Foundry Azure Content Understanding FAQs

We are excited to share some recent updates that improve the experience of using Workload identity federation (OpenID Connect) with Azure DevOps and Terraform on Microsoft Azure. Many working parts have come together to make this possible and we’ll share those here. We are also very pleased to announce version 5 of the Microsoft DevLabs […] The post Introducing Azure DevOps ID Token Refresh and Terraform Task Version 5 appeared first on Azure DevOps Blog.  ( 25 min )

As described in our official document, the classic communication mode of Batch node will be retired on 31 March 2026. Instead, it’s recommended to use simplified communication mode while creating Batch pool.   But while user changes their Batch pool communication mode from classic to simplified and applies the necessary changes of network security group per documentation, they will find out that the node is still stuck in unusable status.   A very possible cause of this issue is due to the bad networking setting of Batch Account.   This blog will mainly talk about why networking setting can cause node using simplified communication mode stuck in unusable status and how to configure correct networking setting under different user scenarios.   Cause: As described in this document, the differ…  ( 23 min )

As described in our official document, the classic communication mode of Batch node will be retired on 31 March 2026. Instead, it’s recommended to use simplified communication mode while creating Batch pool.   But while user changes their Batch pool communication mode from classic to simplified and applies the necessary changes of network security group per documentation, they will find out that the node is still stuck in unusable status.   A very possible cause of this issue is due to the bad networking setting of Batch Account.   This blog will mainly talk about why networking setting can cause node using simplified communication mode stuck in unusable status and how to configure correct networking setting under different user scenarios.   Cause: As described in this document, the differ…

Microsoft Build is just a few weeks away. To celebrate, we’re highlighting resources that will help you get ready for the big event. Explore some of the exciting sessions you can join in-person or online, learn new skills before jumping into live deep-dive sessions, brush up on best practices, and get up to speed on the latest developer tools so you can hit the event ready to take your knowledge (and your applications) to the next level. Connect, code, and grow at BuildIt’s almost time for Microsoft Build! Can’t join the event live in-person? No problem. You can still experience the event streaming live online for free (May 19-22). Watch the keynote, join live sessions, learn new skills, and watch in-depth demos. Join the .NET & C# teams at Microsoft Build 2025Don’t miss this opportunity t…  ( 28 min )

Microsoft Build is just a few weeks away. To celebrate, we’re highlighting resources that will help you get ready for the big event. Explore some of the exciting sessions you can join in-person or online, learn new skills before jumping into live deep-dive sessions, brush up on best practices, and get up to speed on the latest developer tools so you can hit the event ready to take your knowledge (and your applications) to the next level. Connect, code, and grow at BuildIt’s almost time for Microsoft Build! Can’t join the event live in-person? No problem. You can still experience the event streaming live online for free (May 19-22). Watch the keynote, join live sessions, learn new skills, and watch in-depth demos. Join the .NET & C# teams at Microsoft Build 2025Don’t miss this opportunity t…

As AI reshapes how businesses operate and deliver value, many organizations are seeking ways to modernize their infrastructure — quickly, securely, and at scale. With the right tools and support, cloud adoption becomes an empowering step toward innovation and growth. Azure Innovate & Azure Migrate and Modernize were designed to support your entire cloud journey, providing expert guidance, funding, and comprehensive resources all in one place to help you maximize the value of Azure to boost business growth. We’re continuously evolving these offerings to deliver even more value at scale. That’s why we created Cloud Accelerate Factory, a new benefit of Azure Innovate & Azure Migrate and Modernize, built on the patterns and insights from thousands of customer deployments. The Factory provides …  ( 23 min )

As AI reshapes how businesses operate and deliver value, many organizations are seeking ways to modernize their infrastructure — quickly, securely, and at scale. With the right tools and support, cloud adoption becomes an empowering step toward innovation and growth. Azure Innovate & Azure Migrate and Modernize were designed to support your entire cloud journey, providing expert guidance, funding, and comprehensive resources all in one place to help you maximize the value of Azure to boost business growth. We’re continuously evolving these offerings to deliver even more value at scale. That’s why we created Cloud Accelerate Factory, a new benefit of Azure Innovate & Azure Migrate and Modernize, built on the patterns and insights from thousands of customer deployments. The Factory provides …

Table of Contents Abstract Introduction Prerequisites Workstation setup Repository directory contents Terraform variables file Credentials Azure settings Instaclustr settings VNet settings AKS cluster settings Azure NetApp Files settings PostgreSQL settings OpenSearch settings Authorized networks Infrastructure deployment Application deployment Using OpenMetadata Adding a service Adding an ingestion Cleanup Summary Additional Information Abstract This article contains a step-by-step guide to deploying OpenMetadata on Azure Kubernetes Service (AKS), using Azure NetApp Files for storage. It also covers the deployment and configuration of PostgreSQL and OpenSearch databases to run externally from the Kubernetes cluster, following OpenMetadata best practices, managed by NetApp® Instaclustr®. T…  ( 61 min )

Table of Contents Abstract Introduction Prerequisites Workstation setup Repository directory contents Terraform variables file Credentials Azure settings Instaclustr settings VNet settings AKS cluster settings Azure NetApp Files settings PostgreSQL settings OpenSearch settings Authorized networks Infrastructure deployment Application deployment Using OpenMetadata Adding a service Adding an ingestion Cleanup Summary Additional Information Abstract This article contains a step-by-step guide to deploying OpenMetadata on Azure Kubernetes Service (AKS), using Azure NetApp Files for storage. It also covers the deployment and configuration of PostgreSQL and OpenSearch databases to run externally from the Kubernetes cluster, following OpenMetadata best practices, managed by NetApp® Instaclustr®. T…