Daily, sourced write-ups on Claude Code, Codex, Copilot, Cursor, Devin, MCP, and the models behind them: what shipped, what broke, and what it costs you.
New posts on coding agents, their models, and the infrastructure around them.
Gemini 4 Hydrogen costs $1.50/$7.50 per million tokens. Argon and Carbon double on December 1 and 3.8 Flash on January 1. Worked costs and what to change now.
OpenAI's proposed date for ending GPT-5.6 in Cursor is November 12. Who is exposed, what your own OpenAI key still covers, and replacement models at the same price.
Fable 5.2 leads every coding benchmark Anthropic published, at $8/$40 per million tokens. Where it beats Opus 5.5 on cost per task, plan limits in Claude Code, and the Copilot admin policy.
Ten models, 273 real change requests: the best agent meets 70% of acceptance criteria but fully finishes 26.4% of tasks, and the harness alone swings scores by 19 points.
Cursor cloud agents can now build, host, and share web apps. What the four publish checks and the AI review catch, what they miss, and the admin settings to decide first.
Claude Code 2.1.319 backgrounds long MCP tool calls, can deny instead of prompting, changes auto mode behind LLM gateways, and closes a sandbox escape and three permission holes.
METR estimates Anthropic now improves its models about twice as fast, mostly thanks to AI, while staff self-report 12x speedups. What the gap teaches teams measuring AI coding gains.
Conductor Cloud shuts down November 20 after the Meta deal. How to save your workspaces and branches, and where to run parallel Claude Code and Codex agents next.
Pick the layer you're deciding on.
A practical three-way comparison of the editor agents developers actually evaluate now: context quality, pricing predictability, GitHub integration, and review burden.
After Copilot's billing change and Cursor's VS Code dominance, which AI code editor wins for real development workflows? Context quality, pricing, and ecosystem lock-in compared.
For AWS-heavy teams and JetBrains shops, these are the tools Cursor vs Copilot debates miss. An honest comparison of ecosystem fit, security scanning, pricing, and where each breaks down.
Browse posts by the kind of tool.
Cursor, GitHub Copilot, Windsurf, JetBrains AI, Zed, and Replit: the AI built into the editor you write code in all day.
Claude Code, Codex CLI, Aider, Continue, and Cline: agents that run multi-step changes from your terminal, with your tests in the loop.
Devin, OpenHands, and SWE-agent: tools that take a ticket and come back with a pull request for you to review.
Claude, GPT, Gemini, Llama, Mistral, and Hermes, plus MCP, A2A, and the frameworks that connect agents to tools.