Fetching from the wire…
01
02
03
04
05
06
07
08
Top 5 · 2026-03-04 · source-backed
Just one year after coining "vibe coding," Andrej Karpathy officially retires the term for "agentic engineering" — "AI does the implementation, human owns the architecture, quality, and correctness." Simon Willison simultaneously published a multi-chapter Agentic Engineering Patterns guide (497 points on HN, 283 comments) covering TDD with agents, anti-patterns, and cognitive debt prevention. Anthropic's new 2026 Agentic Coding Trends Report adds production data: agents now complete 20 autonomous actions before requiring human input (double from six months ago), Rakuten achieved 99.9% accuracy on 12.5M-line codebase modifications, and Zapier deployed 800+ internal agents with 89% adoption. This is no longer hype — it's a professional discipline with guides, courses, and production tooling. What to do: Read Willison's guide at simonwillison.net/guides/agentic-engineering-patterns/. Adopt the /simplify pre-PR workflow. Structure your CLAUDE.md for progressive disclosure. (The New Stack, Anthropic)
Each link below shares sources, entities, or timing with this story.
Anthropic's definitive report reveals developers use AI in 60% of work but fully delegate only 0-20% of tasks. Production data: Rakuten tested Claude Code on a 12.5M-line codebase (vLLM) — completed in 7 hours with 99.9% accuracy, zero human code contribution during execution....
One year ago today, Andrej Karpathy fired off what he called a "shower thought throwaway tweet" and accidentally named an entire industry. Vibe coding. The numbers since then are staggering: 92% of US developers have adopted vibe coding practices, the AI coding market hit $8.5...
Three independent sources — Simon Willison at the Pragmatic Summit, blog.tedivm.com's coding agent guide, and Anthropic's 2026 Trends Report — converged in March 2026 on the same conclusion: the highest-leverage skill in AI-assisted development is no longer prompt engineering...
On June 9, Anthropic released Claude Fable 5 and Mythos 5 across Claude.ai, Claude Code (CLI and web), and Cowork. The spec sheet: 1M-token context, 128K max output, a January 2026 knowledge cutoff, and pricing at $10 input / $50 output per million tokens. That's double Opus 4...
Simon Willison has been writing software for over 25 years. He's one of the most disciplined, transparent engineers in the Python ecosystem. And yesterday he published an essay admitting he no longer reviews every line of code that Claude Code generates for his production proj...
Simon Willison spent a while taking ChatGPT Work apart and published the map on August 30. Work splits into Work Cloud and Work Local, the latter being the renamed Codex desktop app, at $20/month and up since July 9. He enumerates six capabilities Work has that Chat doesn't, a...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.