Tech Stack

Description

You will develop and advance vision-language-action models for Nuro's onboard Behavior Planning stack, improving safe decision-making in complex driving scenarios. You'll work on multimodal models connecting scene understanding and planning, from research to deployment on real vehicles.

Requirements

• M.S. or Ph.D. in CS, ML, Robotics, AI, or related field
• 5+ years industry/research experience in ML with large-scale model development
• Strong experience with VLMs, multimodal foundation models, VLAs
• Hands-on experience with large-scale model training, fine-tuning, evaluation
• Strong Python skills and deep learning framework experience (PyTorch, JAX, TensorFlow)

Responsibilities

• Develop and advance VLA models for onboard Behavior Planning
• Build multimodal models for safe decision-making in long-tail scenarios
• Train, fine-tune, and evaluate large-scale VLAs using real-world driving data
• Optimize models for onboard deployment (inference efficiency, latency, memory)
• Collaborate with autonomy, data, and infrastructure teams on deployment requirements