Nvidia, Carnegie Mellon, and UC Berkeley have released ENPIRE, a framework that lets AI coding agents run the full loop of teaching robots new skills with no human supervision. Agents running Codex, Claude Code, and Kimi Code pushed an eight-robot fleet to a 99% success rate on tasks including pin insertion, GPU insertion, and zip-tie cutting. Scaling from one robot to eight cut the time needed to master a task by more than half, though the token bill grew even faster than the time saved. A fleet of eight robot arms at Nvidia's GEAR lab spent the past few weeks teaching themselves to insert pins, seat graphics cards, and cut zip ties. The only humans involved were the ones who wrote the paper afterward. The skill came from ENPIRE, a framework detailed in a paper published Tuesday by researchers at Nvidia, Carnegie Mellon University, and UC Berkeley. The system splits the work into two stages: first, a human walks the agent through building a reset routine and a reward function; once those tools exist, the agent takes over completely. Nvidia ran the experiment on eight bimanual robot stations, each with its own hardware, computer, and coding agent. Scaling from one robot to eight cut the time to master Push-T from roughly five hours to two, and pin insertion from more than 90 minutes to about 40.