Researcher, Locomotion

Ngpartners; Palo Alto (CA); Híbrido; USD 140 000 - 210 000; Tempo integral

Resumo da oferta

Ngpartners is seeking a Researcher, Locomotion to advance Asimov from walking to running, learning to recover and maneuver in the real world. You will own policies for robust whole-body locomotion on hardware, not just in simulation.

You will collaborate with hardware, controls, and manipulation researchers, publish open-source workflow, and push the limits of sim-to-real transfer on a humanoid platform.

Qualificações

• Deep hands-on experience with reinforcement learning for continuous control and locomotion.
• Proven track record of transferring learned policies to real robots.
• Strong skills in physics simulators and domain randomization.

Responsabilidades

• Design, train, and ship reinforcement learning policies for bipedal locomotion on Asimov.
• Own the sim2real pipeline end-to-end, building on zero-shot sim2real work so a first run on hardware is never really a first run.
• Push balance, gait, and recovery behavior past the demo stage into something that survives pushes, slips, and uneven ground.
• Build and refine reward design, domain randomization, and training environments in MuJoCo.
• Close the loop between simulation and hardware using real telemetry from real robots, feeding what breaks back into the next policy.
• Work shoulder-to-shoulder with hardware, controls, and manipulation researchers, as whole-body control spans across team boundaries.

Conhecimentos

Reinforcement learning

Continuous control

Legged locomotion

Real-robot deployment

ROS2

Ferramentas

MuJoCo

Isaac Gym

Python

ROS2

Descrição da oferta de emprego

Type: Full-time

Location Type: Hybrid

Department: Brain

ABOUT OUR CLIENT

Our client is an Applied R&D lab building Asimov, an open-source humanoid robot platform, and the full software stack that powers it. Their mission is to make humanoid labor economically viable, turning software into physical labor at scale. They build across the full stack: hardware architecture, locomotion, autonomy, simulation, and infrastructure. They move fast, ship to real robots, and open-source everything they can. If you want your work to matter beyond a paper or a demo, this is the place.

THE ROLE

Our client is looking for a Researcher, Locomotion to push Asimov from walking to running, recovering, and moving through the real world with the confidence of something alive. Asimov 0 was built to learn locomotion and Asimov 1 to learn whole-body control, and you will own the policies that make that motion robust on real hardware. You will not stop at a clip in simulation—you will get your work onto a physical biped and keep pushing until it holds up under contact, disturbance, and terrain not trained for.

WHAT YOU'LL DO

• Design, train, and ship reinforcement learning policies for bipedal and whole-body locomotion on Asimov.
• Own the sim2real pipeline end-to-end, building on zero-shot sim2real work so a first run on hardware is never really a first run.
• Push balance, gait, and recovery behavior past the demo stage into something that survives pushes, slips, and uneven ground.
• Build and refine reward design, domain randomization, and training environments in MuJoCo.
• Close the loop between simulation and hardware using real telemetry from real robots, feeding what breaks back into the next policy.
• Work shoulder-to-shoulder with hardware, controls, and manipulation researchers, as whole-body control spans across team boundaries.
• Open-source what you can and write up what you learn so the broader community can build on it.

WHAT WE LOOK FOR

• Deep hands-on experience with reinforcement learning for continuous control, legged locomotion, or whole- body control.
• A track record of getting learned policies onto real robots, not only into papers or simulators.
• Strong command of a physics simulator such as MuJoCo, Isaac, or similar, including reward shaping and domain randomization.
• Fluency in Python and modern RL tooling, alongside comfort in a ROS2-based control stack.
• A bias for shipping: a drive to see a policy stumble on real hardware this week rather than look perfect in sim next month.
• Clear analytical thinking regarding why a policy fails, not just whether it does.

Nice to Have

• Published work in locomotion, legged robotics, or sim2real transfer.
• Experience with model predictive control or classical locomotion methods alongside learning-based approaches.
• Contributions to open-source robotics or RL projects.
• Experience bringing up new hardware and debugging the gap between a model and a motor.

WHY JOIN OUR CLIENT

You will own locomotion for a humanoid that thousands of developers are building on in the open. Your policies will ship to real robots—not slides—and the community will see them walk. If you want your research to leave the lab and stand up on its own two feet, this is the place.

A NOTE ON AI

You don't need deep AI expertise for every role, but our client expects everyone to be intellectually

curious, drawn to tinkering and discovery, and excited to use AI as a real collaborator in their work. For some roles, AI fluency is a core requirement; when that's the case, it will be explicitly stated in the qualifications. Team members who thrive here don't treat AI as a novelty—they use it to think better and make their work easier for others to build on.

EQUAL OPPORTUNITY AND ACCOMMODATIONS

Our client hires talented people from a wide range of backgrounds. If you're excited about a role but don't meet every bullet point, you are still encouraged to apply. Our client is an equal opportunity employer and does not discriminate on the basis of any legally protected characteristic. Reasonable accommodations are provided during the application process—if you need one, please let your recruiter know.