robot learning · policy training
ABOUT THE ROLE
As the first robot learning hire at a physical AI data and infrastructure company, you will bring policy training and evaluation in-house. You will work with engineering and product teams to demonstrate how synthetic training data affects robot performance and translate findings into product and data requirements.
WHAT YOU'LL DO
- Train and fine-tune robot policies on synthetic data using imitation learning, vision-language-action methods, diffusion policies, or reinforcement learning in simulation.
- Establish baselines on public evaluation suites and compare results using consistent seeds, checkpoints, and protocols.
- Evaluate manipulation and grasping tasks, reporting improvements and failures with rigor.
- Turn experimental results into data specifications covering what to generate, fidelity, and variation.
- Publish and present findings to the robotics community and participate in customer technical discussions.
- Help hire and grow the robotics team as an individual contributor who takes on team-building responsibilities.
WHAT WE'RE LOOKING FOR
- At least 6 years of hands-on robot learning experience, including work in industry and university research labs.
- A master's or doctoral degree in robotics, machine learning, or a related field.
- Experience training and evaluating robot policies in simulation with Isaac Sim, Isaac Lab, MuJoCo, or equivalent tools.
- Experience with policy learning approaches such as VLA fine-tuning, diffusion policy, ACT, imitation learning, behavior cloning, or reinforcement learning.
- Strong Pyt