Andrej Karpathy, a founding member of OpenAI and former Tesla AI executive, has joined Anthropic to work on research and development tied to training new AI models, the company said Tuesday.

Karpathy will work on Anthropic’s pretraining team, which handles the large-scale training runs that help give Claude its core knowledge and capabilities. Axios reported that he will also help launch a new team focused on using Claude to speed up pretraining research.

The move gives Anthropic one of the most recognizable researchers in artificial intelligence at a time when major AI labs are competing for top technical talent, computing power and stronger model performance.

Karpathy was a founding member and research scientist at OpenAI before joining Tesla, where he served as director of AI and led computer vision work tied to Autopilot. He later returned to OpenAI from 2023 to 2024, according to his personal website.

He has also built a large public following through AI education work. Reuters reported that Karpathy said he remains “deeply passionate about education” and plans to return to that work in time.