PHD RESEARCHER / PHYSICAL AI

I make things
that move.

I study intelligent systems in the physical world. I also make beautiful interfaces and open-source software because the tools we use should feel as considered as the ideas behind them.

Web · Robotics · WAM · VLA·Former RoboMaster member
SCROLL TO EXPLORE ↓

WEB, ROBOTICS, WAM, VLA, OPEN SOURCE

01 Now

A SMALL INTRODUCTION

Curious about the gap between thinking and touching.

I am a PhD student researching Physical AI — systems that perceive, reason, and act in the real world. My work lives somewhere between robotics, learning, simulation, and the messy details of making an idea actually run.

When I am away from the lab, I design interfaces, maintain open-source tools, and obsess over the little moments that make software feel calm, clear, and surprisingly nice to use.

01 / embodied intelligence02 / interface craft03 / open source
02 Research

SELECTED PAPERS

Learning to act
in the physical world.

Physical AI: how models perceive, predict, and act. Papers, preprints, and the code behind them.

01 / PAPEROpenWAM
OpenWAM project graphic

OpenWAM ↗

An open, modular research stack for World–Action Model pretraining, spanning infrastructure, controlled studies, and an open pretrained WAM.

AUTHORSYuran Wang*‡ · Siqiao Huang*‡ · Mingleyang Li* · Chenhao Zhang* · Jiaqi Liang* · Weiyang Jin · Yue Chen · Xuemin Chi · Donghao Zhou · Qize Yu · Yu-Kai Wang · Yuhan Rui · Shenzhe Yao · Zhen Yuan · Zhenhao Shen · Kefei Zhu · Zijie Zhu · Ning Gao · Xiaowei Chi · Guanqi He · Shanghang Zhang · Hao Dong · Lin Shao† · Hang Zhao†* equal contribution · ‡ project lead · † equal advising
DATE2026 · arXiv preprint 2609.07398 ↗
03 Engineering
04 Notes

A LINE I KEEP COMING BACK TO

“The best tools disappear just enough for the idea to come through.”

I write about frontend systems, research tooling, and the places where engineering becomes a design problem.

Read the blog ↗