Robot Learning Engineer - Manipulation
Core
Building a robot learning platform on Dana to enable robots to learn industrial manipulation tasks and improve autonomously.
Role type
Senior IC robot learning engineer (manipulation)
Builds
Physical AI platform for robot autonomy in industrial settings
Domain
Robotics + Physical AI + Industrial Automation
Deliverable
production ML models
Required skills
Python, PyTorch, imitation learning, vision-language-action models, diffusion policies, action-chunking transformers, robot kinematics, camera calibration, low-level control interfaces
Preferred skills
bimanual manipulation, force-controlled insertion, tactile sensing, ROS 2, LeRobot, Isaac, MuJoCo, NVIDIA Jetson optimization, reinforcement learning post-training, policy transfer
Technologies
PyTorch, ROS 2, LeRobot, Isaac, MuJoCo, NVIDIA Jetson
Responsibilities
Train and fine-tune manipulation policies from large pretrained models to compact task-specific policies; develop repeatable recipes for industrial tasks like pick-and-place and assembly; deploy policies on edge compute and validate observation processing and control timing; diagnose failures through controlled experiments; measure and improve customer metrics like success rate and cycle time; package recipes and models for reuse across tasks and robots
Seniority
Senior, hands-on IC
