Completed
26:08 - How close are we to self-improving AI?
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Building Super Intelligent Tools - AI Agent Development and Experimentation
Automatically move to the next video in the Classroom when playback concludes
- 1 00:00 - The acceleration of AI viewed through programming agents
- 2 01:46 - Level-setting on AI agents: nomenclature, evals, and experimentation
- 3 03:18 - Experimentation types and processes
- 4 05:30 - Optimizing each step of the experiment loop
- 5 08:10 - Bringing W&B Launch back to solve knotty experimentation problems
- 6 11:50 - Optimizing the research phase in Weave
- 7 14:43 - Can we use AI to automatically improve the experimental loop?
- 8 17:39 - What changes with the resurgence of reinforcement learning
- 9 19:12 - OpenPipe’s Kyle Corbit on building reliable agents with RL
- 10 23:14 - Overcoming the limitation of evals with researcher agents
- 11 26:08 - How close are we to self-improving AI?
