AI Researcher - Reinforcement Learning
Core
Developing and deploying reinforcement learning policies to teach humanoid robots manipulation and locomotion tasks for safe home operation.
Role type
Senior IC reinforcement learning engineer (robotics)
Builds
RL policies for manipulation and locomotion that run reliably on physical humanoid robots in real-world home environments
Domain
Robotics, AI, Reinforcement Learning
Deliverable
production ML models
Required skills
Python, C++, PyTorch, domain randomization, reward shaping, sim-to-real transfer, PPO, SAC, TD-MPC, large codebase management, build tools (Bazel)
Preferred skills
model-based RL, world-model-guided policy learning, imitation learning, learning from demonstration, legged locomotion, dexterous manipulation, contact-rich control
Technologies
Isaac Sim, MuJoCo, PyTorch
Responsibilities
Train RL policies in simulation for manipulation and locomotion, close the sim-to-real gap, build training and evaluation infrastructure, partner with hardware and controls teams to ship skills to production
Seniority
Senior, hands-on IC
