Robot learning veteran: most factory robots are more capable than their tasks
stepjamUK · x · 2026-09-08
The author has spent over a decade in robot learning — PhD at Imperial, postdoc at Berkeley, and building the Dyson Robot Learning Lab. The pattern he sees hasn't changed: most robots on factory floors today are more capable than the tasks they're used for.
Every major step, from early imitation learning to today's VLAs and diffusion policies, was deployable years before most teams picked it up — deployment lags technical maturity.
Related event: Roboticist: Factory Robots Are More Capable Than Their Tasks(2 posts)→
More from Companies & People
- Lila Sciences cofounder Ben Kompa makes MIT TR35 list for autonomous science push — tw_killian · 2026-09-09
- A 4-phase parallel learning roadmap from linear algebra to neural networks — kmeanskaran · 2026-09-09
- Zach Dell on Base Power: meeting AI's energy demands, batteries and vertical integration — espricewright · 2026-09-09
- Cresta signs NFL legend Marshawn Lynch for 'Beast Mode' AI customer service campaign — JaynitMakwana · 2026-09-09
- Dev open-sources AI-generated Microsoft Fabric Rayfin course outline and materials — adnan_hashmi · 2026-09-09
- Asimov Press returns with focus on original biotech reporting and thesis-driven philanthropy — NikoMcCarty · 2026-09-09