Yifan Su (苏逸凡)

I am a Member of Technical Staff at Skild AI in San Francisco, where I work on robot learning. I am an incoming Ph.D. student in Robotics at Carnegie Mellon University, where I will be advised by Deepak Pathak and Abhinav Gupta. I received my B.S. in Computer Science from CMU in May 2025.

My interests sit at the intersection of robotics and machine learning — in particular reinforcement learning for control and manipulation, and planning and coordination for multi-agent systems.

As an undergraduate at CMU, I worked with Professor Jiaoyang Li on multi-agent path finding, focusing on how robot teams can execute plans robustly when delays occur in the real world. I also worked with Deepak Pathak on scaling reinforcement learning for dexterous manipulation and locomotion, and earlier with Prof. Sebastian Scherer and Prof. Micah Corah on a multi-drone reconstruction project at the CMU AirLab.

Selected Work

Robot Parkour — Skild AI

Work from the team at Skild AI on agile whole-body control: a robot performing parkour-style traversal of real-world obstacles.

"PARKOUR!!" — Skild AI, October 2025. Watch on YouTube

News

• Joining CMU Robotics as a Ph.D. student, advised by Deepak Pathak and Abhinav Gupta.

• BTPG-max was accepted to AAAI 2026.

• May 2025 — Graduated from Carnegie Mellon University with a B.S. in Computer Science, and joined Skild AI as a Member of Technical Staff.

• Mar 2025 — Evolutionary Policy Optimization is out on arXiv.

• Feb 2024 — Presented our BTPG paper as an oral at AAAI 2024 in Vancouver.