Anthropic just landed one of the biggest names in AI.
Andrej Karpathy, the OpenAI founding member who was formerly Tesla's director of AI, announced on Tuesday that he had joined Anthropic.
The hiring marks a major get for Anthropic in the AI talent wars. Karpathy famously coined the term "vibe coding" and is prolific in the AI community, frequently publishing lengthy social media posts about the state of the industry that are widely read.
"Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative," Karpathy wrote on X. "I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time."
Anthropic said Karpathy will work on the AI startup's pretraining team, which is responsible for large-scale testing of Claude. Karpathy started this week and sits on the team led by Nicholas Joseph, another ex-OpenAIer who was an early employee at Anthropic, OpenAI's chief rival in the AI race.
"I can't think of anyone better suited to do it — looking forward to what we build together!!" Joseph wrote on X.
Excited to welcome Andrej to the Pretraining team! He'll be building a team focused on using Claude to accelerate pretraining research itself. I can’t think of anyone better suited to do it — looking forward to what we build together! https://t.co/mQb33SNiSW
— Nicholas Joseph (@nickevanjoseph) May 19, 2026
Karpathy's AI roots extend back to Anthropic's chief rival.
He helped launch OpenAI as a founding research scientist, then left for a stint at Tesla as its AI director, leading the Tesla Autopilot computer vision team, before rejoining OpenAI in 2023. He publicly supported OpenAI CEO Sam Altman during his brief ouster, but ultimately left OpenAI again in February 2024 and later started Eureka Labs, an AI-centered education company.
He became an even bigger name among the tech community in early 2025, when he coined the term "vibe coding" to refer to "where you fully give in to the vibes, embrace exponentials, and forget that the code even exists."
Vibe coding has come to illustrate the extent to which non-technically proficient users can still create with the help of generative AI coding tools, including Claude Code. Karpathy has recently said that AI models have advanced to the point where AI agents themselves are writing the actual code, with oversight from a human user, necessitating a new term.
"Many people have tried to come up with a better name for this to differentiate it from vibe coding, personally my current favorite 'agentic engineering,'" he wrote on X in February.
Karpathy is joining Anthropic at a time when the AI lab has reached new heights, thanks to major advancements in the company's Claude Code and Cowork tools and a boost in broader public interest following a tussle with the Trump administration. Anthropic's valuation on secondary markets recently surpassed $1 trillion, surpassing OpenAI.
At the same time, Anthropic's rivalry with OpenAI has reached a fever pitch, with their respective CEOs notably refusing to hold hands onstage during a photo op with other AI leaders. OpenAI CEO Sam Altman recently accused Anthropic of helping to fuel the hate toward him that led to an attack on his house.
