RL Researcher - Real-World Robot Learning

1X; San Carlos (CA); On-site; USD 200,000 - 300,000; Full time

Benefits offered by this job

Medical, dental, and vision coverage

Generous PTO & parental leave

401(k) with company match

FSA/HSA options

Commuter benefits

Disability and life insurance

Employee Assistance Program

Job summary

1X is building humanoid home robots and is scaling RL capabilities for real-world tasks. The Reinforcement Learning team trains policies for manipulation and locomotion across simulation and real-world environments, then deploys them to homes.

Your work will directly affect how NEO performs in daily tasks and adapts to household settings. You will own end-to-end RL development, collaborate with hardware and QA teams, and push the boundaries of sim-to-real transfer and production deployment for

Qualifications

• Strong Python and/or C++ with experience in large codebases and build tools (Bazel or equivalent).
• Proficiency with PyTorch for RL policy training and experimentation.
• Hands-on experience with simulation platforms (Isaac Sim, MuJoCo, or equivalent) for policy training at scale.
• Demonstrated experience training RL policies for manipulation or locomotion tasks, including addressing the sim-to-real gap on physical hardware.

Responsibilities

• Own the full RL pipeline from algorithm development through production deployment for NEO in real-world home environments.
• Collaborate with hardware, controls, data, and QA teams to ship RL-trained skills to production customer sites.
• Advance sim-to-real transfer techniques to narrow the gap between simulation and real-world policy behavior.
• Build training and evaluation infrastructure with standardized benchmarks and regression detection.

Skills

Python/C++

PyTorch

RL algorithms

Simulation platforms

Imitation learning

Model-based RL

Production deployment

Tools

Isaac Sim

MuJoCo

Codebases (Bazel or equivalent)

Job description

1X is building humanoid home robots and is scaling RL capabilities for real-world tasks. The Reinforcement Learning team trains policies for manipulation and locomotion across simulation and real-world environments, then deploys them to homes.

Your work will directly affect how NEO performs in daily tasks and adapts to household settings. You will own end-to-end RL development, collaborate with hardware and QA teams, and push the boundaries of sim-to-real transfer and production deployment for