Do you have a passion for applying machine learning to drive real-world robot behavior? As a Reinforcement Learning Engineer on the Spot Behavior team, you will develop and deploy cutting-edge reinforcement learning techniques to expand Spot’s capabilities in dynamic, real-world environments. You’ll work on a multidisciplinary team tackling high-impact mobility challenges—ranging from terrain traversal and balance to complex locomotion behaviors. This role offers the opportunity to work hands-on with Spot and push the boundaries of legged robot performance.
Day-to-Day Activities:
- Design and deploy reinforcement learning systems to improve Spot’s mobility and robustness.
- Integrate learning-based solutions into Spot’s existing planning and control systems in collaboration with experts across controls, perception, and planning.
- Build and maintain systems that support reliable, scalable, and reproducible RL training.
- Test and debug your work using our in-house fleet of Spot robots.
- Write high-quality, maintainable code in both Python and C++.
- Provide mentorship and technical guidance on ML best practices.
We are looking for:
- Master’s degree or higher in Robotics, Mechanical Engineering, Computer Science, or a related field.
- 3+ years of experience with a proven track record of deploying models on hardware.
- Proficiency in both Python and C++ programming languages.
- Strong analytical and debugging skills.
- Familiarity with modern deep RL toolkits and architectures.
Nice to Have:
- Experience with legged robotics.
- PhD in Robotics, Mechanical Engineering, Computer Science, or a related field.
Read Full Description