Founding Research Engineer, Robot Learning

The next leap toward general-purpose physical AI won’t come from the lab. It will come from robots deployed in real sites, doing real work, failing, and learning from every correction. That’s our bet, and we partner with some of the leading robot foundation model companies to make it happen.

We’ve built highly accurate remote teleoperation that runs alongside the autonomy stack, so every intervention becomes training signal. We’re now starting deployments in real industrial sites.

• Own our robot stack on YAM, UR and Flexiv arms: control, perception, teleop and the handoff between the policy and the human operator.
• Fine-tune robot foundation models (π-series, GR00T, OpenVLA) on our proprietary real-world data.
• Turn interventions from deployments into better policies.
• Build evaluations that hold up in real factories.

• MS or PhD in robotics, ML or CS, with publications at CoRL, RSS, ICRA, NeurIPS or similar venues.
• A strong engineer in Python and PyTorch, at home with real hardware and ROS 2.
• Hands-on experience fine-tuning VLAs, diffusion policies or world models.
• Bonus: experience at a top robotics or AI lab, training foundation models for manipulation, or working on teleop.

• Your models run on real robots in real factories.
• Proprietary real-world data from day one.
• A founding role with real ownership and a direct line to the CEO.