Research Engineer Robot Learning
1X is a Palo Alto-based AI and robotics company developing general-purpose humanoid robots, led by its NEO home robot.
Palo Alto, United States
About 1X
1X develops humanoid robots that live and learn alongside people. Its flagship product, NEO, is a home robot designed to autonomously help with household tasks, supported by the company’s embodied-AI systems.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will develop sim-to-real reinforcement learning training pipelines for humanoid robots. You will train, test, optimize, and debug robot policies in simulation and on physical hardware. You will model physical systems, tune contact parameters, and prepare human motion data for training.
Requirements
- Master’s degree or equivalent in Robotics, AI or Machine Learning, Mechanical or Electrical Engineering, Computer Science, or a related field
- One year of experience in a robotics and machine learning occupation or related role
- Knowledge of rigid-body dynamics or dynamical systems
- Experience implementing production-grade Python reinforcement learning and machine learning pipelines
- Knowledge of reinforcement learning algorithms and sim-to-real robotics approaches
- Experience implementing and debugging real-time C++ software on robots
- Ability to model multi-body hybrid dynamic systems with contact
- Root-cause analysis and robotic-system troubleshooting skills
- Knowledge of forward and inverse kinematics using human motion data
Responsibilities
- Develop sim-to-real reinforcement learning training pipelines for humanoid robots
- Train and test reinforcement learning policies in simulation and on physical hardware
- Optimize training pipelines for performance and speed
- Model robot dynamics, sensors, and contact parameters to reduce simulation-to-hardware gaps
- Collect, clean, and retarget human motion data
- Support testing and debugging of reinforcement learning policies