Agent Harness Breakthrough: Claude Opus Hits 95.5% on ARC-AGI Beating Humans
ycombinator · x · 2026-08-14
Francois Chabard highlights that using a Recursive Language Model (RLM) and Continual Harness, Claude Opus achieved 95.5% on the ARC-AGI 3 public eval, narrowly beating the 95.4% human-expert baseline.
He argues this is irrefutable evidence that so-called "scaffolding" and prompt engineering deserve deeper research attention. Consequently, YC's next Paper Club in Mountain View on Aug 26 will focus specifically on agent harnesses and is currently looking for additional presenters.
Related event: Opus 5 Tops ARC-AGI-3 via Claude Code(7 posts)→
More from coding & agent
- Dev builds MCP server that lets AI agents outsource phone calls to real humans — Many_Benefit_3393 · 2026-10-16
- A vector DB is not LLM memory: long thread maps write/read paths, failure modes and evals — techNmak · 2026-10-16
- Mergelock: open-source MCP server catches semantic merge conflicts git misses across parallel AI agents — ksadayamuthu · 2026-10-16
- OpenUI 1.0 launches open source, turning agent text answers into live interactive UIs — heyshrutimishra · 2026-10-16
- 10 GitHub repos to see what your AI agent really did, from Langfuse to claude-code-log — victor_explore · 2026-10-16
- Mosaic: multi-session agentic RL with hierarchical advantage estimation — LARK-Lab · 2026-10-16