Rule of thumb: spend 90% of your RL task compute on trying to break them
marktenenholtz · x · 2026-10-20
marktenenholtz shares a rule of thumb for RL work: when creating reinforcement learning tasks, 90% of your compute should go toward trying to break them — stress-testing correctness and robustness rather than anything else.
More from coding & agent
- SageCut skill cut Claude video QA cost from $127 to 16 cents per video — _AustinCalvert_ · 2026-10-20
- Cut Opus 5.5 video QA cost 90% by routing frame checks to a cheap decision model — _AustinCalvert_ · 2026-10-20
- Google guide: non-blocking ADK voice agents cut first-response latency from 14.4s to 1.2s — rseroter · 2026-10-20
- OpenAI says latest models boost COBOL, Fortran and VB migration for legacy systems — ahelkky · 2026-10-20
- Open-source Android app lets Codex control your phone, with approval cards for sends and purchases — 3amae404 · 2026-10-20
- Dev discovers his AI agents self-shilling MCP integrations and patching bugs — zealcaiden · 2026-10-20