Kodiak Robotics
World Model Research Scientist- Physical AI
$190K–$250K
Mountain View, CA +1 · Full-time · Posted Mar 27, 2026
- Location
- Mountain View, CA
Bay Area, California, United States - Education
- PhD in Computer Science, AI, Robotics, or a related field.
- Visa sponsorship
- Visa sponsorship may be provided for eligible candidates, if applicable.
About the job
Leads research and development of generative world models that create physically consistent, multi-sensor driving scenarios for autonomous vehicle training, validation, and onboard use.
What you'll do
- Design and train generative models for multi-camera video and LiDAR conditioned on trajectories, 3D scenes, and text.
- Develop conditional diffusion, multi-view consistency, and joint camera, LiDAR, and radar generation methods.
- Create quality evaluation frameworks and scale training on real-world multimodal driving data.
What they ask for
- PhD in Computer Science, AI, Robotics, or a related field, focused on generative modeling, neural rendering, or video synthesis.
- Strong publication record or demonstrated research contributions in relevant generative modeling areas.
- Experience with neural rendering and view synthesis, and understanding of multi-view geometric consistency.
- Proficiency in Python and PyTorch, multimodal sensor data, 3D representations, and distributed training of large generative models.
Pay
$190,000 — $250,000 USD (base salary)
Tools
Python, PyTorch, camera, LiDAR, radar, BEV grids, voxel fields, tri-planes
AI focus
Develop generative world models for physically consistent, action-conditioned, multi-sensor driving scenarios, supporting autonomous-driving training, validation, and onboard models.
Model-written summary · the original posting is authoritative