# Sergey Levine

> Sergey Levine is a UC Berkeley computer scientist and co-founder of Physical Intelligence. After studying computer graphics and machine learning at Stanford, he moved into robot learning and became known for research that connects perception, action and improvement through experience. His current work asks how general-purpose models can help different robots perform useful tasks in changing real-world settings.

- **Role:** Co-founder, Physical Intelligence; Associate Professor, UC Berkeley
- **Organizations:** Physical Intelligence, University of California, Berkeley, Berkeley Artificial Intelligence Research Lab, Robotic Artificial Intelligence and Learning Lab, Stanford University, University of Washington, Google
- **Education:** BS and MS in Computer Science, Stanford University, PhD in Computer Science, Stanford University
- **Known for:** Named to MIT Technology Review's Innovators Under 35 list in 2016., Received a National Science Foundation CAREER award in 2017 and a Sloan Research Fellowship in 2019., Named a recipient of the Presidential Early Career Award for Scientists and Engineers by the NSF's 2025 recipient listing.

## Career timeline

- **2009** — Completed BS and MS degrees in computer science at Stanford University.
- **2014** — Completed a Stanford PhD under Vladlen Koltun and began a postdoctoral appointment in Pieter Abbeel's Berkeley robotics lab.
- **2015** — Worked as a research scientist at Google on machine learning for robotic control.
- **2016** — Held a University of Washington faculty appointment and joined UC Berkeley's EECS faculty in the fall.
- **2024** — Co-founded Physical Intelligence; the company published its first generalist robot policy, π0, in October.
- **2026** — Physical Intelligence reported deployment experiments including coffee service and box assembly at Dandelion Chocolate.

## Achievements

- Named to MIT Technology Review's Innovators Under 35 list in 2016.
- Received a National Science Foundation CAREER award in 2017 and a Sloan Research Fellowship in 2019.
- Named a recipient of the Presidential Early Career Award for Scientists and Engineers by the NSF's 2025 recipient listing.
- Co-authored research on end-to-end visuomotor policies and large-scale robotic grasping.
- Co-founded Physical Intelligence, which released the generalist π0 robot policy in 2024.

## Latest updates

- **2026-08** — Discussed humanoid robotics, generalization and scaling in a public video interview.
- **2026-04** — Physical Intelligence introduced π0.7, a steerable robot model evaluated across multiple robot platforms.
- **2026-02** — Described Physical Intelligence's coffee and chocolate-box deployment experiments in an interview with the Association for Advancing Automation.
- **2025-07** — Published 'Sporks of AGI,' an essay arguing for the importance of real robot experience in training.

## Links

- Website: https://people.eecs.berkeley.edu/~svlevine/
- LinkedIn: https://www.linkedin.com/in/sergey-levine-5a31a24
- Twitter/X: https://x.com/svlevine
- GitHub: https://github.com/svlevine
- YouTube: https://www.youtube.com/@sergeylevine4167

---

Profile page: https://yespress.io/sergey-levine
Published by YesPress — https://yespress.io
Last updated: 2026-09-29