Member of Technical Staff, Robotics Infrastructure
XDOF, San Mateo Hybrid
What you will do
Robotics Infrastructure engineers own the systems that get models running on robots and measure how well they work. Sample projects include:
- deploying and running learned robot policies (VLAs, imitation learning, DAgger) on real hardware, including inference setup, on-robot tuning, and failure diagnosis
- designing and building evaluation frameworks that benchmark policy performance on physical robots, with metric definition, result aggregation, and partner-facing reporting
- building the product that lets partners configure, launch, and monitor evaluation runs and model deployments
- building orchestration systems that schedule evaluation runs and model deployments across robot stations
- integrating LLMs and foundation models into robot workflows for task specification, error recovery, or high-level planning
- containerizing and packaging robot-side software stacks so new stations and deployments can be brought online quickly and consistently
- bridging on-robot systems with cloud-side compute, storage, and orchestration
