Robotics ML Research Intern: Real-World Policy & Evaluation

One Robot (YC W26); San Francisco (CA); On-site; USD 25,000 - 39,000; Full time

Job summary

One Robot (YC W26) in San Francisco is offering an internship to work on real robotics problems, including world model training, policy evaluation on real hardware, and building data/training tooling.

You\'ll embed with the technical team and contribute to active research across world models, policy training, and evaluation, gaining hands-on experience with real robots and simulation environments.

Qualifications

• Currently enrolled in BS/MS/PhD in ML, robotics, CV or related field.

Responsibilities

• Contribute to training runs: work with world model or eval model experiments end-to-end.
• Run real-robot evaluations: collect demonstrations, run policies on hardware, document failure modes.
• Build tooling: write Python and PyTorch code to improve data engine, training pipelines, or eval infra.
• Close the sim-to-real gap: run experiments to test how well simulation predicts real-robot behavior.

Skills

Python

PyTorch

ML basics

Robotics

Education

BS/MS/PhD candidate in ML/Robotics/CS

Tools

Git

MuJoCo

Job description

One Robot (YC W26) in San Francisco is offering an internship to work on real robotics problems, including world model training, policy evaluation on real hardware, and building data/training tooling.

You\'ll embed with the technical team and contribute to active research across world models, policy training, and evaluation, gaining hands-on experience with real robots and simulation environments.