Thoughtworks Tech Radar Vol. 35: Agent Sandboxing Is Now Non-Negotiable
Thoughtworks Technology Radar Vol. 35 is out. Agent sandboxing moves to Adopt, human-in-the-loop theater gets named, and harness engineering goes mainstream.
NVIDIA Cosmos Framework: Post-Train Physical AI in a Day
NVIDIA open-sourced Cosmos Framework with TAO 7 agent skills. Post-train a physical AI world model ...
.NET 11 at .NET Conf: C# 15, Runtime Async, MCP
.NET Conf opens November 10 with .NET 11 GA. C# 15 union types land after ...
Claude Enterprise Inference Hooks: Build the Security Server
Inference hooks are GA for Claude Enterprise. Here is how to build the HTTPS server, ...
Android Studio Rabbit 2: Run Any AI Agent in Your IDE
Android Studio Rabbit 2 brings Bring Your Own Agent to 40+ AI coding tools including ...
Revolut Cut AI Coding Costs 90% With Self-Hosted GLM
Revolut moved 2,000 engineers off Claude Code and Codex to a self-hosted GLM-5.3 stack. Here's the cost breakdown, infrastructure details, and what it ...
Mistral Large 4 Weights Are Out: What It Takes to Run Le Chonk
Mistral released the weights for Mistral Large 4 — a 1.05T-parameter open-weight model nicknamed Le ...
gstack: Turn Claude Code Into a Full Engineering Team
gstack gives Claude Code 23 specialist slash commands — CEO, QA, security auditor, and more. ...
Google Three-Live: Run AI-Generated 3D and AR in 50 KB
Google's Three-Live compiles AI-written Three.js into safe glTF and WebXR. Apache 2.0, 50 KB, works ...
Go 1.27 Upgrade Guide: Generic Methods and JSON v2
Go 1.27 ships generic methods, a stdlib UUID package, and a faster JSON engine. Here ...