No Priors: Andrej Karpathy on Code Agents, AutoResearch, and the Loopy Era

Summary

Andrej Karpathy explores what happens when AI agents can design experiments, collect data, and improve without a human in the loop. The episode covers coding agent capabilities, his AutoResearch project enabling autonomous experimentation, training, and optimization. Karpathy describes the shift from linear AI workflows to loopy ones — agents that iterate autonomously. Key takeaway: we’re entering an era where the bottleneck shifts from human execution speed to the quality of the objectives we set for autonomous systems.

Source: No Priors Podcast Date: 2026-03-20

Listen to episode