Andrej Karpathy, a prominent AI researcher known for co-founding and previously working at OpenAI, as well as leading AI initiatives at Tesla, has officially joined Anthropic.
Confirming his move, Karpathy announced on X on Tuesday, "I’ve joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D."
Karpathy commenced his new role at Anthropic this week, where he is focusing on pre-training under the guidance of team lead Nick Joseph. Pre-training is a critical phase, responsible for the extensive training runs that equip Anthropic's Claude model with its fundamental knowledge and capabilities. The company notes this phase is also among the most resource-intensive and expensive in the development of frontier AI models.
An Anthropic spokesperson informed TechCrunch that Karpathy's primary responsibility will be to establish and lead a new team dedicated to leveraging Claude itself to accelerate research in pre-training.
Karpathy is distinguished as one of the few researchers capable of bridging the theoretical understanding of Large Language Models (LLMs) with the practicalities of large-scale training. His recruitment to build such a specialized team signals Anthropic's strategic belief that AI-assisted research, rather than solely relying on raw computational power, will be key to maintaining its competitive edge against industry leaders like OpenAI and Google.
His career trajectory includes a focus on deep learning and computer vision during his initial tenure at OpenAI, which he left in 2017 to join Tesla. At Tesla, he spearheaded the Full Self-Driving (FSD) and Autopilot programs until his departure in 2022. He then returned to OpenAI for a year before leaving again in 2024 to launch Eureka Labs, a startup envisioned to apply AI assistants to educational challenges.
Updates on Eureka Labs have been scarce since its inception, leaving the renowned researcher's continued involvement with the startup uncertain. Beyond his professional roles, Karpathy has made significant contributions to AI education, including an online course titled "Neural Networks: Zero to Hero," which guides students in building neural networks from scratch, and a YouTube channel where he regularly shares lectures on LLMs and AI.
Reflecting on his broader interests, Karpathy stated, "I remain deeply passionate about education and plan to resume my work on it in time."
TechCrunch has reached out to Karpathy for further comment regarding these developments.
In a separate, yet equally significant, hiring move, Anthropic has also welcomed Chris Rohlf to its frontier red team. This team is tasked with rigorously stress-testing advanced AI models against severe and emerging threats. Rohlf brings over two decades of experience in the cybersecurity industry, having previously worked with Yahoo's esteemed cybersecurity team, "The Paranoids," and more recently spent six years at Meta. He also served as a fellow at Georgetown’s Center for Security and Emerging Technology, contributing to the CyberAI project.
Expressing his enthusiasm in a post on X, Rohlf remarked, "We have a real opportunity in front of us to dramatically improve cyber security with AI. I can’t think of a better company or team to join at this critical moment in time."
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.