AI agents run 24/7 for a month to rewrite Météo-France's AROME weather model
IgorCarron · x · 2026-11-11
A self-described "crazy project" is nearly complete: rewriting Météo-France's AROME numerical weather prediction model from scientific literature alone. For over a month, AI agents have been running 24/7 to gather literature, understand the equations, write code, launch simulations on a supercomputer, and interpret experiment results.
The model is almost complete and already running. The author reports very promising results, with tuning and extensive validation still ahead — a striking example of autonomous AI agents handling research-grade scientific engineering.
More from coding & agent
- Stop burning tokens: a Claude Code orchestrator prompt that routes tasks to cheaper models — TawohAwa · 2026-11-11
- Amazon paper: multi-agent setups burn 1.4-4.7x tokens for small gains; single agent is the default — rohanpaul_ai · 2026-11-11
- Dev shows off building a retro FPS entirely via Codex Mobile — 503 tests passing — Dimillian · 2026-11-11
- Explosion launches Ellf directory: 52 NLP templates plus multi-agent annotation workflow — _inesmontani · 2026-11-11
- Meme: 'Stochastic AI Agent Architecture' Is Just a Markov Chain in Disguise — PrasoonPratham · 2026-11-11
- OpenCode adds Gemini 4 Carbon and Claude Fable 5.2 support — dean_rie · 2026-11-11