Max Robotics — Global Service Robot Directory + US Certification

About

Lilian Weng is a leading researcher in robotics and reinforcement learning, best known for her pioneering work in dexterous manipulation. Her most influential contribution is the development of policies that enable a robotic hand to perform complex, vision-based object reorientation, as demonstrated in her 2019 paper "Learning dexterous in-hand manipulation," which has garnered over 1,588 citations. She further advanced the field by showing that models trained solely in simulation can solve real-world tasks of unprecedented complexity, such as solving a Rubik's Cube with a robot hand—a feat that earned her 2019 paper over 632 citations. This achievement was made possible by her invention of automatic domain randomization (ADR), a key algorithm for bridging the sim-to-real gap. Weng also introduced asymmetric self-play for automatic goal discovery, enabling a single policy to tackle diverse manipulation tasks. Beyond her research, she has contributed to infrastructure with the OpenAI Remote Rendering Backend (ORRB), enhancing simulation fidelity. As a former head of AI at OpenAI, her work has fundamentally shaped how robots learn and interact with the physical world.