The author optimizes a .NET reporting method step by step with BenchmarkDotNet, measuring each change separately. The final version drops from about 1.55 s to 11.9 ms and allocations fall from 33.1 MB to 0.23 MB. It is a useful case for anyone who wants to check AI-proposed optimizations with benchmarks and output comparisons before merging them.
Claude-generated .NET tests can look complete and still miss the regression they are meant to catch. The author walks through tracing assertions back to their inputs, checking mock boundaries, and verifying that a test fails for the intended reason. It is a practical review checklist for anyone accepting agent-written tests into a C# codebase.
The article examines what a .NET AI agent does when its own guardrails fail, such as a throwing guard, a down classifier, a missing caller or miswired middleware. The author deliberately breaks each guard layer and shows how to make the request blocked instead of silently allowed. It ends with a practical checklist for developers building guarded agents.
Part 3 of Scott Galloway's series builds a temporal predictor in C#. It tracks a journey's centroid, velocity and acceleration in real time, then forecasts whether it ends in purchase, cancellation or departure within 60 seconds. The engine learns only from actual endpoints and abstains when evidence is thin, which makes it a concrete example for teams building SQLite-backed behavioural or service-health models.
Microsoft has introduced Work IQ, a workplace intelligence layer that lets AI agents read and act on live Microsoft 365 and Dynamics 365 data. Developers can reach it through an MCP server, an A2A endpoint or a REST API instead of wiring up many Graph endpoints. Every request runs under the signed-in user's Entra ID identity.
Bruno Capuano shows a local .NET pipeline that turns podcast audio into SRT or WebVTT subtitles. Whisper handles speech recognition and timestamps, while the s1-mini model only cleans each cue's text, so timing stays untouched. The post also covers the limits he found with long Spanish recordings.
Copilot Autofix for GitHub Advanced Security in Azure DevOps now runs on an agentic foundation. It proposes fixes as pull requests for code scanning alerts from third-party tools and for dependency vulnerabilities. Your team still makes the final merge decision, so the fixes need to be reviewed like any other change.
MCP C# SDK 2.3.0 brings improvements to HTTP discovery, OAuth, stdio shutdown and the ordering of server tools, prompts and resources. It also includes a source-compatibility change: McpClient.ServerInfo is now nullable. Projects that read this property may see new nullable warnings or fail warnings-as-errors builds.
Part 9 of Taiseer Joudeh's MCP series shows how to attach an existing ASP.NET Core MCP server to a hosted Foundry agent without changing its code. The setup covers a Foundry project, an Entra app registration, a custom MCP connection using OAuth identity passthrough, and a toolbox. It matters for .NET teams that want the agent loop in Azure while user-based authorization still works.
The GitHub Copilot SDK lets you keep GitHub's agent runtime while routing model inference to your own Microsoft Foundry or Azure OpenAI deployment through BYOK. The article walks through a .NET 10 console sample that configures ProviderConfig, authenticates with Microsoft Entra ID or an API key, and sends one prompt.
Nick Cosentino explains how to set up C# code intelligence in Claude Code through the csharp-lsp plugin and where that setup stops being reliable. The article separates the language server, the plugin and the workspace as three layers to verify, and it clarifies what diagnostics and symbol navigation can and cannot prove. It is a practical checklist for .NET developers using Claude Code on C# repositories.
A practical guide to adding a Model Context Protocol server to an existing ASP.NET Core app using the official ModelContextProtocol C# SDK. It covers wrapping existing services as read-only tools, locking the endpoint down with JWT authentication and per-tool authorization, and verifying the surface with an MCP client. Useful for teams that want AI agents to use their current APIs without a rewrite.