Software Engineer, Robot Learning & Interfaces

About the role

We are seeking a Software Engineer to own the end‑to‑end path from data collection to a working robot, including the user interface that drives it. The role focuses on integrating existing robot‑learning libraries into our distributed system and delivering customer‑facing capabilities.

Key responsibilities

• Integrate robot learning libraries (LeRobot, DAgger, RL, ACT/diffusion policies, VLA, Nvidia Isaac) into training and deployment pipelines.
• Wire ML artifacts (Weights & Biases, Hugging Face Hub, Google Cloud Storage) and related tooling into the systems that run on robots.
• Own customer‑facing capabilities from concept to live demo, choosing the right approach and ensuring it works.
• Build and improve teleoperation and data‑collection interfaces.
• Leverage simulation when it accelerates development and switch to real robots when needed.

Required profile

• Proven track record of integrating complex libraries and frameworks into production‑grade systems, including handling failure modes.
• Experience moving research‑level ML code toward production readiness.
• Solid frontend development skills, preferably with React and TypeScript, and ability to ship usable interfaces.
• Understanding of distributed systems and how models fit within larger architectures.
• Comfort presenting demos and building solutions that are ready to showcase.

Required skills

• LeRobot
• DAgger
• Reinforcement Learning
• ACT / diffusion policies
• VLA
• Nvidia Isaac
• Weights & Biases
• Hugging Face Hub
• Google Cloud Storage
• React
• TypeScript
• ROS2
• WebRTC
• GStreamer
• WebXR
• Robot simulation (Isaac, MuJoCo)
• Distributed systems