Karpathy joins Anthropic after two years as independent operator
OpenAI co-founder and former Tesla AI director Andrej Karpathy announced he's joining Anthropic's research team, pausing his Eureka Labs education project.
Andrej Karpathy is joining Anthropic. The OpenAI co-founder, former Tesla director of AI, and creator of widely-used neural network tutorials announced the move this week, saying the next few years on the LLM frontier will be "especially formative" and that he's excited to return to R&D with Anthropic's team. He's pausing educational projects including Eureka Labs but plans to return to them later.
Karpathy spent the past two years as an independent operator—publishing tutorials, building Eureka Labs, and maintaining a reputation as one of AI's most independent voices. His choice of Anthropic over OpenAI, xAI, or Google signals where he believes the most interesting research is happening. The timing aligns with Anthropic's recent policy paper on frontier models through 2028 and what appears to be an aggressive hiring phase for large-scale projects.
The hire is a notable win for Anthropic in the ongoing competition for top-tier AI talent. Karpathy's track record spans foundational work at OpenAI, steering Autopilot's vision systems at Tesla, and teaching a generation of practitioners through his Stanford lectures and YouTube series. His decision to join a research lab rather than continue independent work suggests Anthropic has convinced him it's building something worth the commitment.