Machine Learning Intern (Robotics)

Location: San Francisco, CA; hybrid

Company

Develops autonomous systems for heavy construction equipment, including fully autonomous excavators, to improve job-site safety and accelerate critical infrastructure projects.

What you will do

• Train and evaluate world models using real fleet data collected from autonomous construction machines.
• Run architecture explorations and ablation studies, distinguishing meaningful results from noise.
• Build evaluation harnesses and metrics to measure model quality.
• Collaborate with behavior and controls teams to connect model outputs to planning and control.
• Present findings and help establish a foundation for future model development.

Requirements

• Currently pursuing a BS, MS, or PhD in computer science, machine learning, robotics, or a related field, or equivalent research or industry experience.
• Strong Python skills and hands-on experience training models in PyTorch or an equivalent framework.
• Ability to read research papers and turn them into working implementations.
• Comfort working with messy, real-world data rather than curated benchmarks.
• Sound experimental judgment and the ability to distinguish genuine results from noise.

Nice to have

• Experience with world models, video prediction, or sequence modeling.
• Background in reinforcement learning or imitation learning.
• Prior work with robotics or autonomous vehicle data.
• Familiarity with large-scale training infrastructure.

Culture & Benefits

• Work alongside construction specialists and experienced engineers on physical-world AI problems.
• Contribute to systems deployed on real autonomous excavators rather than only simulated environments.
• Participate in an inclusive workplace committed to equal opportunity.
• Reasonable accommodations are available throughout the hiring process.