Reinforcement Learning & Controls Research Scientist- Spot Behavior
Core
Design, train, and deploy reinforcement learning policies that integrate with Spot's control stack to enable robust locomotion on quadruped robots in real-world environments.
Role type
Senior IC reinforcement learning and controls research scientist
Builds
RL policies and control systems for Boston Dynamics Spot quadruped robots
Domain
Robotics, legged locomotion, reinforcement learning
Deliverable
production ML models
Required skills
reinforcement learning, classical control theory, Python, C++, simulation environments, real-time control loop design
Preferred skills
legged robotics, whole-body control, model predictive control, state estimation, sim-to-real transfer
Technologies
Isaac Sim, MuJoCo, PyTorch, RLlib
Responsibilities
Design and deploy RL systems to improve mobility and robustness; Tune and validate low-level controllers; Build and maintain simulation environments; Analyze robot data logs to diagnose failures; Test and debug directly on the Spot fleet; Write production-ready code in Python and C++
Seniority
Senior, hands-on IC
