Andrej Karpathy’s move to Anthropic is one of the most important talent shifts in artificial intelligence this year. And no, it is not because he is a loud public figure, but because he has worked at three places that define the current AI race: OpenAI, Tesla and now Anthropic.
He wrote on X, ““I think the next few years at the frontier of LLMs will be especially formative,” Karpathy wrote in a post on X, referring to large language models. “I am very excited to join the team here and get back to R&D.”
After helping to start OpenAI, Karpathy joined Tesla in 2017 to serve as director of AI. There, he led the computer vision team for Tesla Autopilot. Musk has repeatedly referred to Karpathy as the #2 guy in the world in computer vision,” behind Ilya Sutskever, another OpenAI co-founder.
Karpathy left Tesla in 2022, briefly went back to OpenAI before starting AI education startup Eureka Labs, where he has worked until now.
He is also the 3rd senior OpenAI figure to defect to Anthropic in the last two years. Jan Leike left in may 2024, John Schulman (co-founder) left in august 2024 and now karpathy.
Anthropic said Karpathy will join its pretraining team, the group that works on the large-scale training behind Claude’s core abilities. His job will be to ensure that Claude can help make itself smarter.
As a firm trying to compete with OpenAI, Google and Meta, Karpathy’s hiring comes at a crucial time for Anthropic. Often known as the GOAT in the artificial intelligence space, Karpathy is expected to bridge the gap between LLM theory and large-scale training practice. Tapping him to build such a team is a clear sign from Anthropic that it believes AI-assisted research, rather than pure compute, is how it stays competitive in the long run.
