Role Summary
Lead the development of Vision-Language-Action (VLA) models that enable humanoid robots to understand instructions, perceive environments, plan actions, and learn from large-scale data.
Responsibilities
· Define overall AI architecture for VLA systems.
· Develop multimodal models combining vision, language, and robot actions.
· Lead training pipelines for imitation learning, behavior cloning, RL, and foundation models.
· Design data collection and annotation strategies.
· Optimize model deployment on edge robotics hardware.
· Collaborate with perception, controls, and simulation teams.
· Drive fleet learning and continuous improvement systems.
Requirements
· MS/PhD in AI, Robotics, Computer Science, or related field.
· 8+ years in AI/ML development.
· 3+ years leading AI teams.
· Experience with:
o PyTorch
o Transformers
o LLMs/VLMs
o Reinforcement Learning
o Imitation Learning
o Diffusion Models
· Strong knowledge of robotics and autonomous systems.
· Experience training models on multi-GPU clusters.
Preferred
· OpenVLA, RT-2, ACT, Diffusion Policy, LeRobot, Open X-Embodiment.
· Experience with humanoid robotics.
Benefits:
Work Location: In person
