Andrej Karpathy is a Slovak-Canadian AI researcher, engineer, and entrepreneur who helped usher deep learning into mainstream practice. After co-founding OpenAI, he built Tesla's Autopilot vision team, returning in 2023 to help refine GPT-4 and ChatGPT's assistant capabilities. In 2024 he founded Eureka Labs to democratize AI literacy through hands-on tools and globally accessible courses.
Teaching Deep Learning To The World
Karpathy designed Stanford's pioneering CS231n course and later launched the popular "Neural Networks: Zero-to-Hero" YouTube series. His lectures, blog posts, and open-source projects such as micrograd translate cutting-edge research into approachable material. These resources have trained tens of thousands of practitioners and set the template for modern AI education.
Scaling Vision For Self-Driving Cars
As Senior Director of AI at Tesla (2017-2022) he led the vision team that replaced radar with camera-only perception and introduced the vector-space architecture showcased at Tesla AI Day 2021. Under his leadership Autopilot's neural networks expanded from highway lane-keeping to city-streets FSD, training on billions of frames streamed from the fleet. His work proved that large-scale data and end-to-end learning could power real-world autonomy.
Democratizing AI Through Eureka Labs
In 2024 Karpathy left OpenAI to launch Eureka Labs, an AI+Education startup aimed at giving anyone the tools to train and deploy powerful models. The company produces interactive curricula, open-source notebooks and evaluation sandboxes that lower the barrier to entry for students and startups. He continues to share insights at conferences such as YC's AI Startup School 2025, advocating for an open, vibrant AI ecosystem.