World Model Research Scientist- Physical AI
Core
Design and train generative world models that synthesize realistic multi-camera video and LiDAR conditioned on ego trajectories, 3D scene context, and text to enable scalable closed-loop training for autonomous trucks.
Role type
Senior IC research scientist (generative world models)
Builds
Generative world models for autonomous driving simulation and validation
Domain
Autonomous driving, generative AI, computer vision
Deliverable
production ML models
Required skills
Generative modeling, neural rendering, diffusion models, video synthesis, multi-view geometric consistency, multimodal sensor data processing, 3D representations (BEV grids, voxel fields, tri-planes), distributed training, Python, PyTorch
Preferred skills
Experience with large-scale generative model training, research contributions in world models or 3D-aware generative models
Technologies
PyTorch, Python, BEV grids, voxel fields, tri-planes
Responsibilities
Design conditional diffusion architectures for driving, develop techniques for multi-view geometric consistency, build methods for joint multimodal generation, design evaluation frameworks for world model quality, scale training pipelines for real-world driving data
Seniority
Senior, hands-on IC