Muse Amazed Me Again — One-Shot Imitation Learning, and Your Taste Is the Ceiling
For ten years, one of reinforcement learning’s cleanest dreams has been a machine that learns a new skill from a single demonstration. This weekend I watched...
I write about applied AI and engineering leadership — an enthusiast attempting to find balance through coding, marathoning, meditation, and fasting.
For ten years, one of reinforcement learning’s cleanest dreams has been a machine that learns a new skill from a single demonstration. This weekend I watched...
The connector this post is about is open source — here it is, steal it: github.com/LiangGou/muse-tesla-connector. Here is how it got built, and why.
For the past several weeks I have been talking with Muse, a personal agent that writes, codes, plans, searches, and reasons alongside me — and it can do a lo...
“Quick decision,” my Muse messaged me, in the middle of a feature build. “Found a small bug in the module next door. Do we fix the bug first — or finish the ...
Try it right now: the app is live at ai-grader-shwa.onrender.com public demo account — username public, password public — with its own isolated...
Inspired by “Coding Models之后” (After Coding Models) by Hokin的学术笔记 (Hokin’s Academic Notes) — original post. [1] An English translation of the original post i...
Part Two of a series inspired by “Coding Models之后” (After Coding Models) by Hokin的学术笔记 (Hokin’s Academic Notes) — original post. [1] Part One covered renderi...
AI is making intelligence abundant.
There’s a debate going around about Jev, the new model from TypeSafe. One camp says it’s exactly what AI needs: a fast, cost-efficient System One to compleme...
I didn’t expect to write this post. But two things happened recently that I can’t stop thinking about, and they both point to the same uncomfortable conclusi...
For years, my thinking lived in design docs, meeting notes, and papers. Useful to my team — invisible to everyone else.