Robotics Researcher

Location: United States (SF Bay Area preferred)

Employment: Full-time

Aiming to be the world’s leading humanoid company, — humanoids that work reliably in the real world every day, built with mass-production discipline rather than demo discipline.

Three vertically integrated components:

- Omakase D1— our own humanoid hardware

- Omakase Zen — the manipulation intelligence foundation (this role)

- Omakase OS — the orchestration software that runs robots in the field

Our moat is data. Through our own fleet and a partnership with a leading spot-labor platform, we have structurally exclusive access to real-world teleoperation and egocentric human data at a scale competitors cannot replicate. Zen turns that flywheel into manipulation foundation models.

 

 The Role

You will train the models that make our robots' hands work.

Today that means post-training VLA policies (π0.5-class, ACT) on our own teleoperation data and deploying them onto real robots doing real jobs in Japan. Next it means pre-training a cross-embodiment manipulation foundation model on our proprietary dataset.

You own the loop end to end: data curation → training → evaluation → real-robot deployment → failure analysis → back to data. We do not have a separate team that "puts the model on the robot." That handoff is where robot learning usually goes to die.

 

 What you would actually work on

Concrete problems currently open on our side:

- Action-space representation. Our end-effector-space policies are systematically weaker on rotation than on translation. Until the representation is fixed, EE-space deployment is blocked. This is an open problem we would want your opinion on in the interview.

- Cross-embodiment transfer. We collect human hand trajectories and retarget them to the robot. How much of that transfers, and in which representation, is not yet settled by our own measurements.

- Evaluation that predicts real-robot success. Open-loop MSE on held-out episodes is cheap and weakly correlated with what happens on the robot. We build sim-based harnesses and structured real-robot trials, and we would like them to disagree less.

- Making failures legible. When a policy misses a grasp, the useful question is which part of the pipeline was wrong — the data, the representation, the training, or the hardware. We invest in being able to answer that.