1X
Research Engineer, Robotic Learning
San Carlos, CA$166.6k–$350kfulltimemidAdded today
About this role
Design and build simulation environments for humanoid robots, developing reinforcement learning algorithms to teach manipulation and locomotion tasks. Bridge the sim-to-real gap by training robust policies and deploying them to production robot fleets operating in home environments.
What you'll do
- Design simulation environments for robot training and policy development
- Train RL policies for diverse manipulation and locomotion tasks
- Implement imitation learning with expert demonstrations and DAgger for covariate shift reduction
- Architect distributed neural network training across GPU clusters
- Close sim-to-real gaps and validate policies on physical robots
- Collaborate with controls, QA, and data collection teams to ship production policies
What they're looking for
- Reinforcement learning and imitation learning
- Generative models (diffusion, VAE) for planning
- Distributed training on GPU clusters (NCCL, data parallelism)
- C++ for real-time systems and multithreading
- Simulation environment design and physics engines
- Policy deployment and robotics systems integration
- Dataset curation and loss function design
- Python for ML/deep learning
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
1X
1X builds humanoid robots for home use, with its NEO robot requiring sophisticated hardware, firmware, and cloud systems. The company is hiring across compliance testing, audio systems, embedded firmware, cloud platform development, and supplier engineering to bring its robots to production and scale.
View all jobs at 1XLikely interview questions
- Describe your experience training large-scale imitation learning models—how did you handle expert demonstration collection and loss function design?
- Walk us through a time you closed a sim-to-real gap in robotics. What were the key challenges?