- Published multiple papers in top conferences like ICRA, ICLR, CVPR, and CoRL.
- Major contributions include:
- DOS®: A Deployment Operating System for Robots
- EMS®: A Massive Computational Experiment Management System towards Data-driven Robotics
- RoboFlow: a Data-centric Workflow Management System for Developing AI-enhanced Robots
- Developed a switch trajectory transformer with distributional value approximation for multi-task reinforcement learning
- Proposed a meta path planning algorithm via neural exploration-exploitation trees for high-dimensional planning
Research Experience
- Interned at Meta Reality Lab, Meta’s Ranking & Foundational AI, and Zebra Tech.
- Involved in multiple robotics projects such as DOS®, EMS®, and RoboFlow.
Education
PhD in Computer Science at Northwestern University, advised by Prof. Han Liu.
Background
Broadly interested in robot learning, focusing on efficiently scaling AI-driven robotics through computational systems and efficient decision-making. Drawn to developing simple, scalable methods that leverage both on-board and distributed compute to accelerate real-world robotic deployment.