Key Takeaway
Andrej Karpathy, a co-founder of OpenAI and former Tesla AI leader, has joined Anthropic’s pretraining team to work on Claude. The move strengthens Anthropic’s technical bench as frontier AI labs compete for elite researchers, compute capacity, and faster model-development workflows.
Andrej Karpathy Joins Anthropic – Key Points
The Story
Andrej Karpathy announced on May 19, 2026, that he has joined Anthropic, where he will return to research and development at the frontier of large language models. Anthropic says Andrej Karpathy will work on its pretraining team, the group responsible for the large-scale training runs that shape Claude’s core knowledge and capabilities. He will also help build a team focused on using Claude itself to accelerate pretraining research.
The Facts
Andrej Karpathy starts at Anthropic this week.
He will work under Anthropic pretraining lead Nick Joseph, focusing on one of the most expensive and compute-intensive stages of building a frontier AI model.
His new work focuses on AI-assisted AI research.
Anthropic plans for Karpathy to build a team that uses Claude to accelerate pretraining research, a sign that frontier labs are increasingly trying to use advanced models to improve future model development.
Andrej Karpathy is one of the most visible AI researchers in the industry.
He was a founding member of OpenAI, later led Tesla’s AI and computer vision work for Autopilot and Full Self-Driving, and has built a large public following through AI education content.
His career links several major AI centers.
Andrej Karpathy worked at OpenAI from 2015 to 2017, joined Tesla as director of AI, left Tesla in 2022, returned to OpenAI for about a year, and left again in 2024 before launching Eureka Labs.
Education remains part of his long-term work.
Andrej Karpathy said he remains “deeply passionate about education” and plans to resume that work in time. Eureka Labs, his AI education startup, has focused on applying AI assistants to learning, and he has also taught neural network and LLM concepts through online courses and public lectures.
The hire comes during a wider frontier AI talent battle.
Anthropic’s decision to put Andrej Karpathy on pretraining highlights how much frontier labs value researchers who understand both LLM theory and large-scale training practice.
Anthropic has also added cybersecurity talent.
Chris Rohlf, a veteran cybersecurity researcher with more than 20 years of experience, has joined Anthropic’s frontier red team, which stress-tests advanced models against severe risks. Rohlf previously worked at Yahoo’s security team, Meta, and Georgetown’s Center for Security and Emerging Technology.
Why This Matters
Andrej Karpathy’s move matters because frontier AI competition is not only about funding, chips, or product launches. It is also about who can attract researchers capable of improving the training process behind major models like Claude, ChatGPT, and Gemini. For users and businesses, this kind of hiring can influence the speed, reliability, and capability of the AI tools they use every day.
This article was drafted with the assistance of generative AI. All facts and details were reviewed and confirmed by an editor prior to publication.
Anthropic expands Enterprise AI Agents across finance, HR, and engineering as software stocks rebound and SaaS disruption fears ease.
Anthropic unveiled dreaming for Claude Managed Agents, letting AI agents review past sessions, update context files, and improve work.
The Anthropic funding round could raise $30 billion at a $900 billion valuation, driven by enterprise revenue and AI infrastructure demand.
Anthropic says Mythos Preview can exploit critical vulnerabilities and remains withheld from public release over cyber and safety risks.
Read a comprehensive monthly roundup of the latest AI news!
