Microsoft Security Blog
2026-11-12
Microsoft describes how poisoned datasets and model checkpoints in the post-training supply chain can compromise coding agents, leading to remote code execution and other malicious behavior. The post is a security analysis from Defender Research and the AI Red Team.
Visual Studio Blog
2026-11-11
Bring Your Own Model (BYOM) is now generally available in Visual Studio 18.12. It lets developers connect hosted models, provider API keys, or local models directly in Visual Studio.
.NET
2026-11-10
The post announces .NET 11 and mentions C# 15, faster developer tools, runtime and library improvements, and new capabilities for web, mobile, desktop, and AI apps.
2026-10-29
Microsoft Threat Intelligence reports that North Korean threat actors are increasingly using AI in their operations, with different levels of adoption across groups. The post describes how AI is being incorporated into attack chains to improve speed and efficiency.
Microsoft Command Line Blog
2026-10-28
The article says rapid AI model development and broader technology maturity are changing what forward-deployed engineering means. It frames the role as evolving over the last 24 months, but provides no operational details in the excerpt.
GitHub
GitHub announced “GitHub intelligence,” a set of features meant to give agents more context about how a team works, not just the codebase. The post says preferences, repository conventions, skills, and custom instructions can influence how changes are made.
2026-10-27
The article says agentic software delivery extends beyond code generation by connecting people, AI, and feedback across the software development lifecycle.
2026-10-15
Microsoft and Carnegie Mellon University suggest that junior developers trained today will become the senior engineers working alongside AI-assisted coding tools in the future. The piece frames AI as part of software engineering training rather than a replacement for developers.
Azure Updates
2026-10-08
Microsoft Marketplace multiparty private offers are now expanding to Hong Kong. This lets partners procure third-party cloud and AI solutions for customers in the Hong Kong market through Marketplace.
2026-10-07
GitHub announced a new purpose-built, context-aware model for leaked secret detection as part of Secret Protection. It is intended to improve secret detection in code workflows, including those that involve AI agents.
Let's do DevOps
The article describes a toolchain using Whisper, FFmpeg, and yt-dlp to download sermon audio/video, transcribe it, and analyze the text for potentially bigoted or otherwise problematic language. It frames the approach as using AI to review sermons for “red flags.”
2026-10-06
The post says CISOs should adapt vulnerability management practices for AI-driven environments to reduce cybersecurity risk and improve resilience. It frames the topic as managing vulnerabilities in the age of AI-powered tooling and threats.