ML Engineer II, Navigation (Robotics)

TL;DR

ML Engineer II (Navigation): Develop learning-based navigation models that enable service robots to move naturally and safely around people, beds, wheelchairs, and equipment with an accent on imitation learning from fleet data and RL refinement in simulation. Focus on building pipelines for trajectory extraction, reward shaping, domain randomization, and integrating policies into safety systems for real hospital deployments.

Anywhere in the US

Company

We build artificial intelligence that enables service robots to collaborate with people and adapt to dynamic human environments in hospitals.

What you will do

• Develop learning-based navigation models that predict safe, smooth trajectories from sensor inputs and perception representations.
• Build imitation learning pipelines from fleet logs including trajectory extraction, filtering, scenario balancing, and evaluation.
• Implement simulation-based refinement using RL, reward shaping, and domain randomization to improve robustness.
• Define navigation success metrics aligned to product outcomes like delivery speed and reduced interventions.
• Collaborate with AI Platform team to integrate policies into behavior and safety systems and validate on-robot.
• Analyze field behavior, identify failure modes, and iterate through data curation and retraining.

Requirements

• Bachelor’s or Master’s degree in Robotics, Computer Science, Electrical Engineering, or related field (PhD a plus).
• 3+ years of experience in ML for robotics, autonomy, or sequential decision-making.
• Strong proficiency in PyTorch and experience with sequence models / policy learning.
• Experience with imitation learning and/or reinforcement learning in robotics or autonomy contexts.

Nice to have

• Experience with socially-aware navigation and dynamic obstacle avoidance.
• Experience with RL at scale including simulation rollouts, distributed training, and stability/debugging.
• Familiarity with ROS navigation stacks and safety constraints for mobile robots.
• Experience building eval harnesses like offline replay and scenario libraries.
• Experience with Vision-Language-Action (VLA) models, behavior cloning, and/or transformer/diffusion policies for robotic control.