Linas Nasvytis

I am a PhD student at Stanford, advised by Judith Fan and Noah Goodman. I’m currently spending the summer as a Research Fellow at Goodfire, working on interpretable self-improvement.

My research focuses on the intersection of human and machine intelligence. I am particularly interested in experience-efficient reasoning – how intelligent systems can quickly and continuously learn from minimal data – and its applications for automated scientific discovery.

Before Stanford, I earned a master’s in statistical machine learning at Oxford, working with Jakob Foerster, Christian Schroeder de Witt, and Chris Summerfield. I also spent a wonderful year at Harvard collaborating with Sam Gershman and Fiery Cushman.

Before that, I was working with Paul Romer at NYU on economic research, tech policy, and cryptography. I graduated from the Dual Bachelor’s progam between Columbia University and Institut d’études politiques de Paris.