AI
Build Your Own AI Agent Harness in C#, the MafClaw Live Series
I am building a complete C# agent live, from a single call around an IChatClient to a production-ready, observable, governed agent, using the Microsoft Agent Framework harness in a 4-part Microsoft Reactor series.
Your AI coding agent evaluation is only as good as its sandbox
Your AI coding agent passed the eval. But did the model know the answer, or did it find it somewhere on your machine? A correct answer can still invalidate your measurement.
Build an interview coach app with the GitHub Copilot SDK
An interview coach has to do more than ask questions. It needs to read a resume, follow up on an incomplete answer, and save enough context to give useful feedback at the end. Some of that work is conversation. Some of it requires calling an application service. The GitHub Copilot SDK lets you use the runtime behind Copilot CLI for that work inside your own application. You provide instructions and callable tools. Copilot handles the model interaction and resulting tool calls, while your application owns the interface and business workflow. For a developer building a personal assistant or an internal workflow, ...
Coding agents are picking Azure SQL Database
I went looking for Azure SQL Database in a recent coding agent benchmark, honestly unsure whether Microsoft SQL would show up. Armature ran a study where real coding agent CLIs (Claude Code, Codex, Cursor) were dropped into six synthetic-but-realistic repositories and asked to pick a database, then actually build with it. 356 runs were completed across three personas: vibe coder, junior developer, senior engineer. Every session is published, so you can replay the prompts, the searches, the commands, and the code. Azure SQL Database came in second! Not second behind AWS. Second behind Neon, and ahead of Amazon...
Building a Japanese LLM Evaluation Pipeline: Lessons from a Two-Day Hackathon
ISE hackathon produced a reusable Japanese LLM evaluation pipeline with multi-turn, facet-based scoring, cross-provider model support, and publication-ready benchmarking outputs.
Advancing trusted AI innovation with Agents and new Azure OpenAI models in Azure Government
Today, we are announcing expanded AI capabilities available across the Microsoft Azure for US Government portfolio. The Microsoft Foundry Agent Service is now available in Azure Government, giving organizations managed infrastructure on which to build, deploy, and operate AI systems that can reason over authorized information, use approved tools, and help complete multi-step work. Additionally, GPT-5.6 Sol, GPT-5.6 Terra, and GPT-5.6 Luna are now available through Microsoft Foundry in Azure Government. Together, these capabilities give public-sector organizations more choice in how they build secure, scalable AI ...
Your work might not need the smartest model
The smartest model can cost five times more and deliver the same result, or even a worse one. See how evaluating your own work helps you get more value from your agent budget.
How Uno Platform uses .NET, MCP, and AI to build high quality apps
How Uno Platform built two Model Context Protocol servers in C# so AI agents can ground themselves in real documentation and then see and drive a running cross-platform .NET app.
New Azure SQL Foundations video series with GitHub samples
Bob Ward and I recently released a series of videos, read more in the original blog post or go directly to the series on YouTube. The Azure SQL Database Foundations series are four videos that take you from your first Hyperscale database to AI features running against your own operational data. We also included how to assess and migrate (with AI and skills!) to Hyperscale in the first place, and the common optimizations you should consider. Every episode ships with a repo, so you can follow along in your own environment instead of watching someone else’s terminal.
From dotnet run to Foundry Hosted Agent in 3 lines of C#
You built an agent with Microsoft Agent Framework and it works great on your machine. Now what? Turns out deploying it to production takes 1 NuGet package, 3 lines of C#, and 2 commands. Let's do it.