About the role
- The Senior Reinforcement Learning Engineer is a key, hands-on role focused on achieving state-of-the-art performance on our humanoid robots
- This engineer will leverage their deep expertise in RL to solve critical locomotion and manipulation challenges and deliver breakthrough results on physical hardware
- The primary focus of this role is to rapidly implement, iterate, and deploy advanced learning algorithms to push the boundaries of what our robots can do
- As a senior member of the team, this individual will also be responsible for mentoring junior engineers, elevating the team’s overall technical capabilities through their guidance and expertise
- Implement and deploy state-of-the-art RL algorithms to achieve ambitious, world-class performance on dynamic locomotion and manipulation tasks with physical hardware
- Drive the entire development cycle, from prototyping in simulation to robustly transferring and fine-tuning policies on the robot
- Optimize and scale the RL training pipeline for faster iteration, contributing to core infrastructure for high-throughput simulation and distributed training
- Mentor junior engineers by providing technical guidance, conducting insightful code reviews, and sharing best practices in reinforcement learning and software development
- Collaborate closely with the robotics and hardware teams to diagnose system-level issues and co-develop solutions that enable more complex learned behaviors
- Analyze and present hardware results to guide future technical directions and demonstrate progress on key company objectives
- Develop and refine motion retargeting pipelines to translate human demonstration data (mocap, teleoperation) into robust reference trajectories for reinforcement learning
- Must be able to lift 15 pounds at times
- A strong intuition for robot dynamics and controls theory, with the ability to apply these principles to guide and constrain learning-based approaches
- Deep, hands-on expertise (5+ years) with common RL frameworks (e.g., PyTorch, JAX) and high-fidelity physics simulators (e.g., MuJoCo, IsaacGym)
- Hearing and speech to communicate
- A results-oriented mindset with a passion for seeing complex algorithms work on real-world hardware
- A strong theoretical understanding of modern reinforcement learning, including deep expertise in areas like imitation learning, model-based RL, and sim-to-real transfer techniques
- Experience building or utilizing large-scale, distributed training pipelines and a strong intuition for their optimization
- Prolonged periods of sitting at a desk and working on a computer
- A PhD or MS in Computer Science, Robotics, or a related field, with 2+ years industry experience strongly preferred
- A strong publication record in relevant conferences or journals (e.g., CoRL, RSS, ICRA) is a significant plus
- Mastery of Python for rapid prototyping and training, alongside strong proficiency in C++ for developing performant, deployable code
- Demonstrated experience mentoring or providing technical guidance to other engineers in a team environment
- Vision to read printed materials and a computer screen
- A proven track record of successfully deploying learning-based policies on physical robotic systems, especially legged robots or manipulators