YC-backed Reific launches RL environments for time-series reasoning, frontier models score under 50%
ycombinator · x · 2026-11-11
YC-backed Reific launches RL environments for superhuman reasoning over time, where data streams in on a clock and agents are graded by what actually happened.
- Three task families: Timestep-Assist (personal agents on phone data), Timestep-Ops (wind farms, refineries, factories), and Timestep-Lab (bioreactors, GPU clusters).
- Leaderboard: best frontier models average under 50% per family — Gemini 4 Argon 31.6% on personal agents, Claude Opus 5.5 leading industry (43.1%) and science & compute (29.7%).
- Training an open model on these environments raised full-solve rate on unseen tasks from 25.6% to 39.5% (+54%) with no loss on general benchmarks.
More from coding & agent
- Spotify's Claude Code 90% token-saving trick put to a real test — aakashgupta · 2026-11-11
- Bolt: Gemma 4 runs 100% on-device on iPhone to drive a robot dog health companion — DynamicWebPaige · 2026-11-11
- Dev shows off live voice-call AI agent 'Elise' that answers family calls and triages urgency — bigaiguy · 2026-11-11
- agents.md isn't just for AI agents — humans can grok codebases fast with it too — k7agar · 2026-11-11
- Animate real places in 3D with Claude and free map data from AWS and Overture — drbronk · 2026-11-11
- Microsoft Foundry pattern: version and deploy Toolboxes separately from agents behind stable endpoints — WirelessLife · 2026-11-11