nostalgebraist: coding agents are brilliant yet still blunder constantly
JacquesThibs · x · 2026-10-27
LessWrong blogger nostalgebraist argues that today's LLM agents are impressive in many ways yet keep making basic mistakes. Coding agents went from "not useful enough" at the start of the year to net-useful thanks to speed and obscure knowledge, but blunder rates haven't dropped. He jokes the experience is an ego boost — the genius model still can't match him — and muses that it's increasingly unclear what co-working process with agents will actually work well.
Related event: Blogger: Coding Agents Are Smart but Keep Making Basic Mistakes(2 posts)→
More from coding & agent
- Alan AI launches to stop developers babysitting their coding agents — Scobleizer · 2026-10-27
- Compound Engineering ships ce-sweep: turns Slack/GitHub feedback into an executable dev plan — kieranklaassen · 2026-10-27
- Mintlify: doc traffic up 10x in a year, agent traffic up 135x, 60% of edits agent-written — handotdev · 2026-10-27
- Training Agents series wraps at GitHub SF: 4 talks on post-training open-model agents — SergioPaniego · 2026-10-27
- Grok Bot usage guide: event-based triggers and clear prompts beat routine wake-ups — XFreeze · 2026-10-27
- Dev builds Netlifyi, a Grok bot that digests Netlify changelogs daily at 9am — thisiskp_ · 2026-10-27