Senior Scientist, Safe Legged Locomotion & RL

Amazon; Pasadena (CA); On-site; USD 167,100 - 226,100; Full time

Benefits offered by this job

Health insurance

RSUs

401(k) matching

Paid time off

Job summary

Amazon Robotics Compass Team seeks an Applied Scientist to own safe legged locomotion algorithms and their deployment on physical hardware. You will train policies for quadrupeds and humanoids using RL, with sim-to-real transfer, integrated with model-based control to ensure stability and safety in dynamic terrains.

The role emphasizes collaboration with perception, planning, and safety teams, publishing results, and delivering robust autonomous locomotion across varied environments at Amazon’s

Qualifications

• PhD, or Master’s with 6+ years of applied research experience.
• 3+ years applying RL to physical robotic systems (beyond simulation).
• Demonstrated expertise in sim-to-real transfer for locomotion or manipulation tasks.
• Strong understanding of legged robot dynamics, contact mechanics, and whole-body control fundamentals.
• Proficiency in Python and deep learning frameworks (e.g., PyTorch, JAX) with experience building custom RL pipelines.
• Experience with physics simulators for robotics (e.g., Isaac Gym/Sim, MuJoCo, PyBullet).
• Track record of publications at top-tier venues (e.g., RSS, CoRL, ICRA, NeurIPS, ICLR, IROS).

Responsibilities

• Design, train, and deploy RL policies for dynamic legged locomotion on physical platforms.
• Collaborate with safety teams to enforce safety constraints during/after training.
• Develop sim-to-real transfer pipelines robust to reality gap.
• Integrate learned policies with model-based whole-body controllers.
• Formulate reward structures balancing performance and safety constraints.
• Develop large-scale training infrastructure including simulation environments and parallel training.
• Evaluate policy performance via simulations, hardware experiments, and failure analysis.
• Explore emerging techniques and assess their applicability to safe locomotion.
• Publish research and contribute to Amazon’s robotics reputation across domains.

Skills

Reinforcement learning in robotics

Python & DL frameworks

Sim-to-real transfer

Legged locomotion dynamics

Publications & research

Education

PhD or Master’s with 6+ years research

Tools

Isaac Gym/Sim

MuJoCo

PyBullet

Job description

Amazon Robotics Compass Team seeks an Applied Scientist to own safe legged locomotion algorithms and their deployment on physical hardware. You will train policies for quadrupeds and humanoids using RL, with sim-to-real transfer, integrated with model-based control to ensure stability and safety in dynamic terrains.

The role emphasizes collaboration with perception, planning, and safety teams, publishing results, and delivering robust autonomous locomotion across varied environments at Amazon’s