Safe RL Control Engineer

Description

• Ensure controllers are modular, deterministic, and extensible, supporting both classical and learning-based control strategies.
• Define clear, robust interfaces between classical controllers and learned components, enabling smooth blending and fallback behaviors.
• Validate controllers in simulation and hardware environments, iterating closely with system-level testing teams.
• Collaborate with the Imitation Learning and Deployment teams to ensure compatibility of runtime systems and deployment pipelines – while maintaining full ownership of control and WBC components.
• Benchmark actuator properties (like torque limits and delays) to refine simulation models, closing the sim2real gap.
• Develop and enforce safety mechanisms within WBC to manage contact, stability, and recovery during combined locomotion and manipulation (loco-manipulation) behaviors.
• Shape RL action spaces to promote safe exploration, avoiding extreme behaviors while enabling smooth policy execution.
• Design, implement, and extend whole-body control frameworks that coordinate multiple robot subsystems (locomotion, manipulation, teleoperation).
• Architect and tune low-level controllers for balanced performance, supporting compliant behaviors for learning tasks and precise fallback modes for safety.
• Participate in design reviews, profiling, and performance analysis for high-impact control modules.
• Work with deployment teams to align RL outputs with hardware realities, using simulation penalties and transfer techniques for reliable rollout.
• Develop and maintain mid-level controllers that translate motion objectives into coherent, stable, real-time control actions.
• Collaboration with top‑tier engineers, researchers, and product experts in AI and robotics.
• Collaborate daily with control engineers across Boston, London, and Vancouver, aligning control strategies, architecture, and codebase.
• Develop and integrate RL-based controllers and policies within the WBC architecture.

Requirements

• Demonstrated ability to collaborate across geographically distributed teams and disciplines.
• M.S. or Ph.D. in Robotics, Control, Mechanical Engineering, Computer Science, or related field.
• Experience developing or integrating reinforcement-learning-based control policies for high-DOF systems.
• Experience validating control architectures both in simulation and on physical hardware.
• Background in real-time or distributed control systems, including ROS2 or real-time middleware.
• Strong software engineering skills: modular design, benchmarking, testing, and performance profiling.
• Freedom to influence the product and own key initiatives.
• Deep understanding of robot dynamics, kinematics, and control optimization.
• 5+ years of experience developing control software for complex robotic systems (humanoids, legged platforms, or articulated manipulators).
• Strong theoretical and practical background in classical control (model-based control, observers, optimal control, QP-based control).
• Proven ability to design and implement real-time control algorithms in C++ or Python.
• Familiarity with whole-body control frameworks, including task hierarchies, optimization-based control, and constraint handling.

Benefits

• Comprehensive health insurance coverage.
• Paid vacation with adjustments based on your location to comply with local labor laws.
• Travel opportunities to our London and Vancouver offices.
• Competitive salary plus participation in our Stock Option Plan.

Training + Development

Interview process

Visa Sponsorship

Security clearance