YC Decoded podcast: how coding agents are generalizing into robot-use agents
ycombinator · x · 2026-09-26
YC's new Decoded episode features the founders of Waddle Labs and Robocurve discussing robot-use agents: general-purpose models that can control different robots, write policies, and learn physical tasks with little robot-specific training.
- Framed by a recent essay from MIT professor Philip Isola arguing general models may win in robotics
- Covers the lineage from code-as-policies and vision-language-action models to general LLMs, plus the Bitter Lesson applied to robotics
- Discusses harnesses and evals as a form of robot intelligence, with a live demo of Astra controlling a robot
- Closes on how close we are to general-purpose robots
Related event: YC Podcast Explores Coding Agents Generalizing into Robot-Use Agents(4 posts)→
More from coding & agent
- Andrew Ng: rigid testing on early-stage AI projects is a recipe for stall — DeepLearningAI · 2026-09-26
- Peter Steiberger says he codes with Codex plus an OC harness, not Claude — steipete · 2026-09-26
- Can AI audit my 1000-video music library? A practical video-understanding use case — SkipMorrow · 2026-09-26
- One prompt makes Claude generate a 2-min video game history film entirely in code — prasenx · 2026-09-26
- Ogre MCP: open-source MCP server relays live game state to your AI agent, WoW Classic first — bttf88 · 2026-09-26
- Run /doctor prompt-audit to find stale paths and conflicting instructions in Claude Code — daniel_mac8 · 2026-09-26