Andrej Karpathy

Andrej Karpathy is an AI researcher and the founder of Eureka Labs, an initiative dedicated to modernizing education in the age of artificial intelligence. Through Eureka Labs and his public work, he advocates for a future where AI-driven tools transform how people learn, create, and explore new ideas.

Over the past decade, Karpathy has played a central role in shaping the modern AI landscape. He was an early member of OpenAI, contributing to the development of large-scale deep learning models and helping define the organisation’s early research direction. He later joined Tesla, leading AI development and overseeing the company’s efforts in computer vision and autonomous driving. After several years at Tesla, he returned to OpenAI, helping to build and launch a model integrated into ChatGPT in 2023.

Beyond his research and industry leadership, Karpathy is a prominent voice in AI education. His lectures, talks, and online content have reached millions worldwide, including his one-hour lecture on deep learning, which has garnered over 3.1 million views.

Karpathy’s academic journey began at the University of Toronto, where he earned a double major in Computer Science and Physics with a minor in Mathematics. He completed a Master’s degree at the University of British Columbia, developing learning controllers for physically simulated figures. He then earned a PhD at Stanford University under Fei-Fei Li, working alongside Daphne Koller, Andrew Ng, Sebastian Thrun, and Vladlen Koltun during his first-year rotations. While at Stanford, he designed and taught the groundbreaking CS231n: Convolutional Neural Networks for Visual Recognition course, which became one of the university’s most popular and influential AI classes.

Karpathy also gained practical experience through internships that shaped his understanding of machine learning. In 2011, he joined Google Brain to work on large-scale unsupervised learning from videos; in 2013, he interned at Google Research, focusing on large-scale supervised learning on YouTube videos; and in 2015, he completed an internship at DeepMind.