Research Scientist, Vla Models - Atlas

Boston Dynamics

Waltham, Massachusetts, USA
Large-scale behavior models
Multimodal input
Grounded actions
Architect and train end-to-end VLA and Large Behavior Models for mobile manipulation on Atlas

Job Summary

  • Architect and train end-to-end VLA and Large Behavior Models for mobile manipulation on Atlas.
  • Build large-scale imitation learning pipelines that learn from human demonstrations, teleoperation, and simulation data.
  • Deploy, debug, and iterate your models directly on physical hardware.

Matching Summary

Architect and train end-to-end VLA and Large Behavior Models for mobile manipulation on Atlas.

Skills & Requirements

Must-have

  • large-scale behavior models
  • multimodal input
  • grounded actions
  • imitation learning pipelines
  • hierarchical skill systems
  • PyTorch, JAX

Nice-to-have

  • deploying learned policies on real robots
  • curating large demonstration datasets
  • RL as complementary tool
  • publications in top-tier conferences

Key Requirements

  • MS with 3+ years or PhD
  • prior experience training and deploying learned policies
  • strong background in behavior cloning
  • strong background in large behavior models
  • strong background in multimodal learning

Work Rights

Not specified

Tailored Resume

Cover Letter