Research Scientist: Pretraining

The role

You will build the base intelligence layer for robotics by training large-scale robot foundation models. This involves designing and running core large-scale training efforts using massive multimodal datasets.

What you will do

• Designing and executing large-scale pretraining runs for robot foundation models
• Defining model architectures, objectives, and training curricula across multimodal robotic data
• Developing scalable data mixtures and sampling strategies across petabyte-scale datasets
• Guiding data collection operations towards new directions and sourcing new datasets
• Running ablations to understand scaling laws, data quality effects, and architecture tradeoffs
• Collaborating closely with ML Infra and Systems to push cluster utilization, throughput, and reliability
• Turning raw robotic interaction data into generalizable model capabilities

Required qualifications

• Senior/staff (L6+) experience working on large-scale ML systems
• Deep experience training large transformer or diffusion models at scale
• Experience leading or significantly contributing to multi-node, multi-GPU distributed training efforts
• Experience with scaling laws, optimization dynamics, and large-model failure modes
• Strong PyTorch fundamentals
• Comfort debugging at every layer of the stack

About the company

Generalist is on a mission to build general intelligence for the physical world. The team builds embodied foundation models with a focus on dexterity.