Machine Learning Scientist/Engineer
Core
Design and build foundation models for general-purpose robots (mobile manipulators, dual-arm platforms, humanoids) that perceive, reason, and act in the physical world using large-scale video and robot-interaction data.
Role type
Senior Research Scientist (Robotics Foundation Models)
Builds
Novel model architectures, pre-training objectives, post-training methods, and scalable data/training systems for embodied AI agents.
Domain
Robotics, Embodied AI, Multimodal Machine Learning
Deliverable
research
Required skills
Multimodal foundation models, Scalable training (multi-node, large datasets/models), Top-tier research publications (ICRA, CoRL, CVPR, NeurIPS, ICML, ICLR), Modern ML frameworks, Distributed training pipelines, Imitation learning, Reinforcement learning, Self-supervised learning
Preferred skills
Experience translating research to deployed capabilities, Collaboration with robotics/hardware teams
Technologies
PyTorch, Distributed training infrastructure, Large-scale video/robotics datasets
Responsibilities
Design and run experiments on model architectures and learning objectives; Build scalable pre-training/post-training methods; Curate and evaluate large robotics datasets; Develop distributed training pipelines; Partner with engineering teams to validate real-world performance; Communicate findings via publications and internal reviews.
