FrontierCS 2 launches on Kaggle: 48 tasks make agents beat published CS papers
MeganRisdal · x · 2026-10-27
FrontierCS 2, a benchmark for AI agents in CS research, is now live on Kaggle in partnership with Google DeepMind.
- 48 tasks built from published papers across 13 CS areas, each refined with domain experts
- Each task deletes the paper's contribution from released code; agents must outperform it on the authors' own metrics
- Constraints: no paper access, no internet, 5-hour limit
- Organizers note plenty of headroom remains; results and leaderboard are public
Built by @wenhaocha1's team and co-launched with Kaggle and Google DeepMind.
More from coding & agent
- BELL: Open-Source Stack Combines Browserbase, Eve, iMessage and Stripe Link for Personal Agents — jeff_weinstein · 2026-10-27
- Legora runs 2,000 parallel agent evals on Modal Sandboxes for legal AI — blelbach · 2026-10-27
- Two engineers rebuilt a FedEx supplier's billing core in 3.5 months vs an estimated 7-8 — alex_verem · 2026-10-27
- GPT-6.1 Terra scores 48 on AA Coding Agent Index at one-third the cost of GPT-5.6 Terra — ArtificialAnlys · 2026-10-27
- Shift-1 router cuts coding agent spend up to 55% while keeping SWE-bench accuracy — ypatil125 · 2026-10-27
- Non-coder vibe-codes a Godot JRPG with Claude Code in two weeks, music and art synced — ADHDavid · 2026-10-27