Human Interactive Driving Intern – World Models

Responsibilities

Conduct original research in one or more areas: world modeling, multi-agent interaction, reinforcement learning, perception, or simulation-to-reality transfer.

Collaborate closely with full-time researchers on the design, training, and evaluation of learning-based driving systems.

Contribute to building and experimenting with task-aware, multi-modal, and uncertainty-aware models.

Develop and evaluate prototypes in closed-loop simulation environments and, time permitting, on high-performance autonomous driving hardware.

Present research findings through internal talks and work towards a top-tier academic publication.

Integrate and work with large-scale datasets (open-source and internal).

Qualifications

Currently enrolled in a Ph.D. program in Computer Science, Robotics, Machine Learning, or a related field.

Strong background in machine learning, particularly in areas such as deep learning, generative models, reinforcement learning, or probabilistic modeling.

Demonstrated experience with one or more of the following: World models (e.g., latent dynamics, diffusion-based models), Model-based RL or decision-making, 3D perception or sensor fusion, and Large-scale simulation for robotics or autonomous systems.

Prior publication(s) in top-tier conferences (NeurIPS, ICLR, ICML, CVPR, ICRA, CoRL, etc.)

Proficiency with Python and PyTorch.

Familiarity with AWS services (S3, EC2, and SageMaker) and open-source driving datasets (nuScenes, Waymo, Argoverse, etc.) is a plus.