Vision‑Language‑Action (VLA) Research Intern
Core
Designing and training world-action models (autoregressive/diffusion-based) to predict future sensor observations or latent states conditioned on ego actions, integrating them with planning for end-to-end or model-based driving policies.
Role type
Research Intern (Vision-Language-Action Models for Autonomous Driving)
Builds
Scalable autonomy research projects and world-action models
Domain
Autonomous Driving / ADAS / Generative AI / World Models
Deliverable
research
Required skills
generative models (diffusion, autoregressive transformers, VAEs), world models / video prediction, end-to-end autonomous driving, deep learning frameworks, reinforcement learning, model-based planning
Preferred skills
top-venue publication record (CVPR, ICCV, ECCV, ICLR, NeurIPS, ICML), driving benchmarks (NavSim, Bench2Drive), large-scale distributed training, multi-modal/video data pipelines
Technologies
diffusion models, autoregressive transformers, VAEs, deep learning frameworks
Responsibilities
Designing and training world-action models, Integrating world models with planning, Summarizing research findings in papers or patents
Seniority
Intern