Goldberg: harnesses retain edge only in tail cases; niche domains will keep them
yoavgo · x · 2026-10-04
Expanding on his harness distillation argument, Yoav Goldberg notes real harnesses may still outperform models trained to imitate them, but mainly in edge cases—likely not worth the operational cost for common scenarios. His assessment applies to well-resourced mainstream domains; niche verticals will continue to rely on dedicated harnesses.
More from coding & agent
- Replit CEO Amjad Masad: general models should JIT-train their own smaller replacements — amasad · 2026-10-04
- Startup claims it will 'kill a $1T industry'; agent dev Jason Kneen mocks the approach — jasonkneen · 2026-10-04
- Running Claude-dev + Codex-review loop, but can't wire Gemini Pro in as reviewer — bocondo · 2026-10-04
- New edition of 'An Introduction to Programming Languages' adds browser runtime and AI chapters — alfcnz · 2026-10-04
- pg-jev: an open-source Postgres extension for querying rows in plain language — Born_Excuse_5610 · 2026-10-04
- New research prototype tests whether defensive mechanisms can stop autonomous web agents — Admin-ABC-XYZ · 2026-10-04