We just released a free course: GitHub Copilot app for Beginners. It includes 7 chapters and a hands-on project that walks through core agent-development concepts.

The @github engineering team built a best-in-class, truly multi-model harness, which we've implemented across Microsoft products.

Introducing Microsoft Copilot harness in Foundry Agent Service, a simpler way to build agents without rebuilding the execution layer.

It handles ontext, and controls so you can focus on your agent’s skills and outcomes.

Watch @jeffhollan's demo to see it in action.

You built it. Now put it in play.

It’s not too late to submit your Inside the Game: Developer Hackathon project by Oct. 27 for the opportunity to have your work recognized and showcased.

Finish your submission: msft.it/6014axj40

#GitHubUniverse kicks off next week, October 28–29.

Grab your pass to join us in person or sign up to tune in virtually.

➡️msft.it/6013aappD

You have an idea. Now take it beyond the scoreline. ​

Build something that changes how football is understood, told, or experienced with AI.

Knowing when to use AI and when to question its output is part of learning to build.

Watch @shanselman, software engineer Julia Gao, and CMU’s Dr. Michael Hilton discuss PRAISE and how to support the next generation of developers.
youtube.com/watch?v=BglzGR…

Classification is a key use case for decision models.

Check out how accurately and quickly Microsoft-Decision-1 can categorize a variety of queries compared to GPT-6 Sol.

Microsoft-Decision-1 achieved the highest accuracy in our 36-benchmark comparison, spanning nearly 150,000 questions across benchmarks kept blind from training.

The challenge starts now.

Build an AI-powered solution that turns synthetic football data into match intelligence, narratives, recaps, or personalized fan experiences.

Register by Oct. 20.

Submit by Oct. 27: msft.it/6018aUKYq

Today, we announced the next step in the Project HydraFusion vision: local models and intelligent routing for GitHub Copilot on Windows.

@cinnamon_msft shows what this can mean for your everyday coding workflows.

Grounding ties extracted values to source evidence. Confidence scores flag uncertainty. A developer-defined JSON schema gives your app structured output.

Together, they help you validate results or route them for review before your app or agent acts.

In Azure Content Understanding, OCR and layout models preserve text, tables and reading order in Markdown before an LLM extracts fields.

The perception layer reads the page. The LLM reasons over that representation.

Our tests showed LLMs outperforming specialized models on contracts and agreements, where meaning lives in the words.

On tax forms and mortgage applications, where layout carries meaning, our existing approach still won.

Match the approach to the document.

An agent reported 46.1 dB for F-03’s ORL at 1310 nm.

That cell was empty. It borrowed the value from the 1550 nm column after extraction flattened the table.

For your document pipeline, preserving structure is part of preserving meaning.

WSL containers API

Access functions to run Linux containers programmatically in your native Windows apps – unlocking scenarios like running local AI workloads or using cloud-based containerized applications locally.

WSL containers CLI

wslc.exe to directly build, run and deploy Linux containers on Windows, or use its built-in alias container.exe to run the same familiar container commands

WSL containers is now generally available.

Build, run and deploy Linux containers directly on Windows with the WSL containers CLI and API 🧵

Go beyond the scoreline.

Inside the Game: Developer Hackathon is open for registration.

Build with AI, football data, and Microsoft technologies to reimagine how the game can be understood and experienced.

Register by Oct. 20: msft.it/6016aY5LY

Welcome to MCP Live!

Join us for a half-day of MCP, from how it’s being adopted across the developer ecosystem to hands-on sessions on MCP at @github, building MCP servers with @code, and more.

youtube.com/watch?v=uydwDk…

Who spent all the tokens? 👀

TokenOps gives multi-agent workflows one shared, run-scoped budget, with policies that can step in before the next model or tool call executes.

The result is real-time control over spend and agent behavior while the run is still happening.

OpenClaw went from a weekend project to one of @github's fastest-growing open source projects.

Six months in, @steipete and several maintainers share what they learned about scaling contributions, trust, code review, and security.

youtu.be/5VSwaUXtPIE?si…

it's.....McBeautiful....

Register and watch live: aka.ms/mcplive/99/x

Join us September 9 for MCP Live!

Dive into sessions on MCP at @github, building MCP servers with @code, using Toolboxes for a unified MCP layer in Microsoft Foundry, and more.

Mark Russinovich used AI-assisted coding to bring ZoomIt from Windows to macOS in just two days.

See how he translated ZoomIt’s core capabilities to a new platform and turned an ambitious cross-platform port into a working, open-source release.

Under the hood, there’s a lot to dig into.

Agent Terrarium runs on a Rust simulation engine at 20 Hz, with React + Canvas handling the view layer. It supports GitHub Copilot and OpenAI-compatible models, with configurable prompts, awareness levels and working directories.

You ars, themes and gear through JSON with hot reload, while agents react to world events and take on file-based tasks.

Need some weekend project inspo?

One of our devs built Agent Terrarium with GitHub Copilot.

It’s an open-source desktop app where AI agents live, move, react, interact and hang out in a tiny world on your desktop.

Merge the latest ready pull request to land it and every unmerged layer below it in one single operation.

Your existing branch protections and required checks still govern what reaches main

Review individual layers in parallel while the stack map keeps the full change in context.

You and your teammates can each review different layers in parallel without blocking further work.

Work with stacks directly on GitHub, the GitHub CLI, the GitHub mobile app, or with a coding agent such as GitHub Copilot using the gh-stack skill.

Start with a branch and pull request for your first change.

Stacked pull requests break large changes into small, reviewable pull requests.

They’re an ordered series of pull requests that each represent focused layers of your change.

TypeScript creator @ahejlsberg puts TypeScript 7 through its paces, compiling the 2.3M-line @code codebase in under five seconds, and breaking down how the new 10x faster Go-powered compiler and language service scales with shared-memory concurrency.

Your dev environment is temporary.

Jimothy is forever.

And if you use VS Code Dev Containers, you can plug WSL containers into that workflow too.

WSLc support is available in the VS Code Dev Containers pre-release. To try it, open Dev Container settings, find “Docker Path,” and set it to: wslc

Then bring Linux containers into your Windows apps.

The new WSL container API lets Windows apps use Linux containers as part of their app logic, with support for C, C++, and C#.

Think local cloud app workflows, Linux-specific code, or contained processes.

Start with the command line.

With wslc, you can run Linux containers as part of your everyday dev workflow, from running apps locally to testing GPU access.

There’s also a built-in container.exe alias, so you can use either wslc or container.

WSL container is now available for public preview.

That means you can create, run, test, and debug Linux containers directly through WSL on Windows.

Here’s how to get started + a few ways to try it 🧵

.@intel and Microsoft are helping developers build more securely in the age of AI with Intel TDX and confidential computing in Azure, designed for sensitive data and workloads.

We partnered with MIT Technology Review Insights to analyze 101 agent use cases and identify the tasks where builders report the highest confidence across data, AI, and cloud workflows.

.@Qualcomm and Microsoft are working together to help developers build for the age of AI agents, with deeper integration across Windows ML, native app experiences, and inference workflows.

.@nvidia and Microsoft are helping developers rethink how they build with AI, from the new NVIDIA N1X chip to working with larger models locally before scaling to the cloud.

Microsoft Build gave us a glimpse of where we’re headed, from Project Solara to the new GitHub Copilot app and so much more.

Thank you to everyone who joined us. Let’s go build together.

Modern DevOps is built on automation and integration. At Build, we showcased a Microsoft partner, Octopus Deploy, which integrates with Azure and GitHub Actions to help teams streamline deployment workflows and ship software through Microsoft Marketplace.

Build, scale and monetize with Microsoft Marketplace.

Scale your app or agent to millions of users through Marketplace and in-product experiences, with built-in global commerce that lets you focus on building, not selling.

Today's challenge is not just creating AI capabilities, it's finding them. We're introducing the Agentic Resource Discovery (ARD) specification, an open spec that establishes a secure common layer for publishing, indexing and discovering AI capabilities. Created by Microsoft, d many more industry collaborators, it's available today to everyone.

Web IQ is Microsoft's state-of-the-art grounding service for AI agents and assistants. It returns ranked, citation-ready context across web, news, images, video, and more. Designed for direct injection into an LLM's context window.

.NET Day on Agentic Modernization kicks off today with expert demos, actionable strategies, and valuable resources to future-proof your business.

Register: msft.it/6011vcxsv

Meet the Majorana 2, a next-generation topological quantum chip developed with the help of Microsoft Discovery’s agentic AI.

Your terminal, now with context-aware intelligence. Debug faster, run multi-step tasks, and stay in flow.

Intelligent Terminal brings your favorite agents into the command line so you can debug, execute, and keep building without breaking focus.

We announced a lot at Build, so consider this your crash course.

Codename MDASH is an agentic vulnerability discovery and remediation system built to help teams find issues, move faster on fixes, and create secure code from the start.

Coming soon to GitHub Copilot CLI and the Microsoft Defender portal.

We met up with @cassidoo at GitHub’s Ship & Tell lounge to ask what she recommends building with right now.

Microsoft Build may be over, but the building continues.

Thanks to everyone who joined us, tuned in, and got hands-on with what’s next.

Important message from our devs at #MicrosoftBuild

During the #MicrosoftBuild keynote, @steipete joined us on stage to talk all things OpenClaw and Windows. 🦞

Now in public preview: Windows 11 developer configuration image for Windows 365, giving developers a preconfigured environment with Visual Studio Code, Git, GitHub CLI, Python, Node.js, and WSL available from first sign-in.

With Microsoft Dev Box in maintenance mode, Windows ing path for teams standardizing developer environments on Cloud PCs.

During the #MicrosoftBuild keynote, @cassidoo took the stage to demo the new GitHub Copilot app, showing how agents, multi-model reviews, custom UI canvases, and enterprise-ready deployment through Rayfin come together in one workflow.

developers, developers, developers

Watch @shanselman and Samantha Song demo OpenClaw on Windows at #MicrosoftBuild.

In the full walkthrough, you’ll see how the Windows Companion app helps you set up your own claws, connect to existing agents, and run OpenClaw across Windows or WSL.

A lot dropped for Windows at #MicrosoftBuild.

Here are @cinnamon_msft’s top 3 updates for developers.

From yesterday’s #MicrosoftBuild keynote: a developer-optimized Windows 11 experience built to help you build and ship faster.

Windows Developer Configurations - powered by WinGet, sets up a distraction-free dev environment with VS Code, GitHub Copilot, WSL, PowerShell 7 and developer-optimized settings with one command on any Windows 11 device.

Intelligent Terminal - intentionally brings context-aware intelligence to your favorite agents directly into a terminal based experience to help debug errors and run multi-step tasks so you can stay in your flow

WSL containers - a built-in way to create, run and interact with Linux containers using familiar CLI & API

Coreutils for Windows - a set of Linux-like command line utilities that run natively on Windows

Say hello to unmetered intelligence on Windows with new on-device SLMs, Aion 1.0 Instruct and Aion 1.0 Plan, and the expansion of Windows AI APIs to more Windows 11 PCs across CPUs and GPUs.

Day 1 of #MicrosoftBuild is officially underway.

Follow this thread for the biggest announcements and moments from the day 🧵

Giving you a reason to say, “That was definitely on my bingo card,” while watching the Microsoft Build livestream over the next two days.

’twas the night before #MicrosoftBuild

We went to #MicrosoftBuild’s backstage event at @github HQ to ask attendees what they’re most excited for ahead of the event.

What are you looking forward to most?

1 day until #MicrosoftBuild

Microsoft Build Live launches tomorrow at 9:30 a.m. PT with the keynote livestream, real-time updates, and developer context as the news unfolds.

w3'11 s33 u th3re

T-5 days until #MicrosoftBuild

Watch the opening keynote and follow top announcements with Microsoft Build Live: your real-time destination for the news, updates, and developer context across Microsoft.

Bookmark it now: msft.it/6014vZMgn

Can't make it to #MicrosoftBuild in-person? We've got you.

Get free hands-on experience remotely with Digital Labs.

Register now: msft.it/6012vkI0Y

The countdown begins! Prepare for #MicrosoftBuild with these tips and start exploring the Build CLI now.

Register for the online experience June 2–3: msft.it/6017vkMIf

Introducing the #MicrosoftBuild CLI.

A GitHub Copilot CLI skill that brings the Build session catalog to your terminal.

Browse sessions by topic, language, or speaker, get recommendations based on your project dependencies, and scaffold working code from what you learn.

#5 The Tools

The capabilities available to the agent, such as reading files, running commands, and searching.

#4 The Prompt

This is the message you send with your goals, requirements, and instructions.

More specific prompts lead to better results.

#2 The Harness

In @code, the GitHub Copilot Chat panel is the harness. It can read your files, run terminal commands, and apply edits directly to your code.

Ask: chat only, no actions
Plan: outlines steps, waits for approval
Agent: Plans, executes, and iterates autonomously
, then move to Agent when ready to build.

#1 The Model

GitHub Copilot lets you choose a model from the model picker, along with how deeply it thinks.

You can choose between Low, Medium, High, and Auto depending on your desired outcome.

#MicrosoftBuild online is free, which means your couch has conference access.

Watch the keynote live, jump into hands-on labs, save sessions for later, and replay what you missed on demand.

Register today: msft.it/6019vr6tP

If you’ve been looking for a sign to attend Build, this is it.

Join us June 2–3 in San Francisco or online for free.

🔗 Register: msft.it/6016vRVCu

GitHub Copilot CLI is even more powerful when you know the right commands.

Save this cheat sheet to speed up your workflows from the terminal.

If your desktop looks like this... it's time to register for #MicrosoftBuild

June 2-3, 2026 / San Francisco & free online.

Register: msft.it/6010vMiyn