Agents made 800k+ production writes with zero human approvals, failure rate 28%→6.5%
TimothyDuignan · x · 2026-08-28
Orbital Industries shares an engineering postmortem on CurieOS: since April, agents inside CurieOS have made 800,000+ writes to production with zero human approvals, cutting task failure from 28% to 6.5%.
When they first let agents write to the database, it was a mess — agents spent time guessing the expected data structure, making up fields or leaving them out. The article details the architecture that made the setup safe and reliable, centered on giving agents explicit structural constraints rather than free rein.
More from coding & agent
- OpenPresence: A Framework for Deploying Autonomous Social Media Agents — RichardsonDx · 2026-08-28
- Can AI Computer Use Agents Fix the 'Notification Void' in Emails? — Darpinian · 2026-08-28
- Chrome extension auto-converts web forms to WebMCP tools — pvncher · 2026-08-28
- Sentry engineer: Clients like Codex are unreliable, so servers must self-heal — zeeg · 2026-08-28
- Trying free Claude Code: proxy routing burns quota instantly, Kaggle self-hosting too slow — phantom_root · 2026-08-28
- Agents Feared Failing METR's Auto-Scorer for 'Cheating' to Grab the Flag — BLUECOW009 · 2026-08-28