Job Description

About Menlo

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

The Role

We are building the motion intelligence that lets Asimov walk, recover, climb stairs, and carry loads without falling over. As a Robotics Researcher in Locomotion, you will work on the Cyclotron team -- Menlo's locomotion training pipeline -- developing the controllers and learned policies that run on physical bipedal hardware. You will train in simulation, close the sim-to-real gap, and deploy to the robot. The bar is real-world robustness, not benchmark performance.

What You Will Do

• Research, develop, and iterate on locomotion controllers and motion policies for a bipedal humanoid
• Train and evaluate policies in Uranus, Menlo's in-house simulation engine, across a wide range of behaviors including walking, recovery, stair climbing, and load-bearing
• Design reward functions, curriculum schedules, and training infrastructure that produce policies robust enough for real-world deployment
• Drive systematic sim-to-real transfer and hardware iteration
• Integrate locomotion outputs with the broader Asimov autonomy stack
• Collect and analyze hardware telemetry to guide policy improvement
• Contribute to open-source releases of locomotion research and Cyclotron tooling
  

What You Will Bring

• Strong foundations in reinforcement learning, optimal control, and rigid body dynamics
• Hands-on experience training and deploying locomotion or motion control policies on physical legged robots
• Proficiency in Python; strong experience with JAX or PyTorch
• Experience with physics simulation environments such as MuJoCo, Isaac Gym, Genesis, or equivalent
• Practical track record closing the sim-to-real gap on a real platform
• Ability to iterate fast, instrument failures, and make data-driven improvements
  

Nice to Have

• Prior work specifically on bipedal or humanoid locomotion
• Experience with whole-body control, model predictive control, or loco-manipulation
• Familiarity with motion capture or real-time state estimation pipelines
• Publications at RSS, ICRA, CoRL, or equivalent venues
  

Why Join Menlo

This is applied robotics research with real stakes -- your code runs on a physical humanoid. We open-source aggressively, so your contributions reach the broader community. You will work alongside researchers and engineers across the full stack, in a team that values shipping over presenting. Competitive compensation and equity.