Top AI Product

Andrej Karpathy announced on May 19 that he’s joined Anthropic — starting a team focused on using Claude itself to accelerate pre-training research. He’s working under pre-training team lead Nick Joseph and started this week.

## The career arc

Karpathy co-founded OpenAI, left in 2017 for Tesla (where he led Full Self-Driving and Autopilot), returned to OpenAI for a year, then left again in 2024 to start Eureka Labs, an AI-for-education startup. The Anthropic move pulls one of the field’s most recognizable researchers — and arguably its best technical explainer — back into frontier-lab research.

## The thesis behind the hire

The team’s mandate is the tell: use Claude to accelerate pre-training research. Anthropic is betting that AI-assisted research, not pure compute scaling, is how it stays competitive with OpenAI and Google. Karpathy building that team signals where Anthropic thinks the next edge comes from — models helping build better models.

## Why it matters

The “AI accelerating AI research” loop is the most consequential bet in the field right now. Putting Karpathy — who has both deep pre-training expertise and a sharp instinct for what’s actually tractable — on exactly this problem is Anthropic’s strongest talent statement of 2026.