Researcher, Locomotion
Core
Design, train, and ship reinforcement learning policies for bipedal and whole-body locomotion on the Asimov humanoid robot platform, ensuring robust motion on real hardware under contact, disturbance, and uneven terrain.
Role type
Senior IC research engineer (legged locomotion & sim2real)
Builds
Robust locomotion policies and sim2real pipelines for the Asimov humanoid robot
Domain
Robotics, legged locomotion, reinforcement learning
Deliverable
production ML models
Required skills
reinforcement learning for continuous control, legged locomotion, whole-body control, physics simulation (MuJoCo, Isaac), reward shaping, domain randomization, Python, ROS2
Preferred skills
published work in locomotion/legged robotics, model predictive control, open-source robotics contributions, hardware debugging
Technologies
MuJoCo, Isaac, ROS2, Python
Responsibilities
Design and train RL policies for bipedal locomotion; own the sim2real pipeline end-to-end; push balance and recovery behaviors to survive real-world disturbances; build training environments and refine reward design; close the loop between simulation telemetry and hardware feedback; collaborate with hardware and controls teams; open-source research findings
Seniority
Senior, hands-on IC
