The Applied Scientist role at the SAF Lab focuses on safe reinforcement learning and the development of legged locomotion algorithms for dynamic robots. The team aims to integrate safety into robotic systems, enabling them to operate alongside humans. This position involves collaboration with various teams at Amazon to advance the field of safe autonomy. The successful candidate will contribute to foundational research and practical applications in robotics.

• PhD in Computer Science, Robotics, Mechanical Engineering, Electrical Engineering, or a related field with a focus on reinforcement learning, robot learning, or control.
• Experience applying reinforcement learning to physical robotic systems, including expertise in sim-to-real transfer on dynamically stable robots.
• Strong understanding of legged robot dynamics, contact mechanics, and whole-body control fundamentals.
• Proficiency in Python and deep learning frameworks such as PyTorch or JAX.
• Experience with physics simulators for robotics like Isaac Gym/Sim, MuJoCo, or PyBullet.
• Knowledge of safety-critical control, including control barrier functions and safety filters.
• Familiarity with safety-constrained reinforcement learning methods.
• Experience with model-based control and its integration with reinforcement learning.
• Knowledge of stability theory as it applies to periodic gaits.
• Experience with hierarchical reinforcement learning and multi-task policy architectures for locomotion.
• Familiarity with real-time deployment constraints.
• Experience building or contributing to large-scale reinforcement learning training infrastructure.
• Strong communication skills and ability to work across disciplinary boundaries.

• Collaborate with product teams and science leaders to set a science roadmap impacting real robots.
• Design, train, and deploy reinforcement learning policies for dynamic legged locomotion.
• Develop sim-to-real transfer pipelines that produce robust policies.
• Integrate control-based methods with reinforcement learning.
• Develop and maintain large-scale training infrastructure for locomotion policy learning.
• Investigate the distillation of locomotion policies and integration with whole-body control.
• Evaluate policy performance through simulation benchmarks and hardware experiments.
• Publish research at top-tier robotics and ML venues.
• Collaborate with perception and planning teams to enable terrain-aware locomotion behaviors.

• Employees at Amazon are often offered comprehensive health benefits—including multiple medical plan options (no pre-existing condition exclusions, 100% covered in-network preventive care), dental and vision plans, a 24/7 medical advice line from day one, expert second-opinion services, and broad mental-health support with several free counseling sessions (including pediatric). Financial wellness typically includes a 401(k) with company match (up to 2%), Restricted Stock Units (equity), FSAs, an emergency savings program, product and partner discounts, and even college-savings and home-purchase programs. Overall, the package is designed to support employees and their families’ health, finances, and day-to-day life.