2027 Internship Behavior Machine Learning Engineer, World Models

The role

You will train and evaluate world models on real fleet data from autonomous machines. You will also run architecture explorations and partner with behavior and controls teams to ship models onto real excavators.

What you will do

• Train and evaluate world models on real fleet data collected from autonomous machines in the field
• Run architecture explorations and ablations
• Build evaluation harnesses and metrics
• Partner with the behavior and controls teams to connect model outputs to planning and control
• Present findings and provide a foundation for the team

Required qualifications

• Currently pursuing a BS, MS, or PhD in computer science, machine learning, robotics, or a related field or bringing equivalent research or industry experience
• Strong Python and hands-on experience training models in PyTorch (or equivalent)
• Ability to read a paper and turn it into a working implementation
• Comfort with messy, real-world data
• Sound experimental judgment understanding the difference between a real result and noise

Nice to have

• Experience with world models, video prediction, or sequence modeling (e.g., DreamerV3, V-JEPA 2, Genie 3, GAIA-2, NVIDIA Cosmos, etc.)
• Background in reinforcement learning or imitation learning
• Prior work with robotics or autonomous vehicle data
• Familiarity with large-scale training infrastructure

Before you apply

• Designed strictly for students (undergraduate or graduate)
• Opportunities for full-time conversion are highly limited

About the company

Bedrock retrofits heavy equipment like excavators to be autonomous. The team includes veterans from Waymo, Segment, and Uber Freight.