About the job
In this role, you will collaborate closely with researchers and engineers on the Learned Behavior teams to tackle plan generation challenges in autonomous driving. You’ll apply state-of-the-art generative modeling techniques—ranging from cutting-edge diffusion, flow matching, energy-based models, and SoTA algorithms—in order to develop novel solutions that generate safe, comfortable, and efficient driving behaviors in the most challenging real world situations. Beyond core research, you’ll own the end-to-end lifecycle of your models, productizing them for robust, real-world autonomous driving deployments on a global scale.
Responsibilities
Develop and scale state-of-the-art generative models—especially diffusion architectures, flow-matching techniques, and energy-based models —for autonomous plan generation.
Build generative models with foundation models. Leverage large language models and world foundation models for reasoning, decision making and multi-modality generation.
Optimize generative models using reinforcement learning to improve interactive reasoning. Explore reward modeling/learned verifier using generative models. Explore joint prediction and planning and self-play. Leverage generative models for active learning and world modeling.
Develop controllable generative models to guide the generation process towards desired goals, conditions and rewards.
Collaborate across autonomy teams while developing holistic solutions to top autonomy challenges. Understand issues, propose ideas, prioritize work and develop solutions to solve them, evaluate your solution by deploying the models on to the NuroDriver.
Qualifications
Minimum
You have a Ph.D. (preferable) or M.Sc. with 3+ years of experience working with generative models in the lab, in industry, or both.
Research experiences in generative models, particularly diffusion models, flow matching and energy-based models, for robotics, including manipulation, path planning, and autonomous driving. Experiences in vision-language-action models, reinforcement learning for generative model optimization, video generation, text-to-image generation, diffusion models for LLMs, world foundation models, and other applications of generative modeling are a bonus!
You have strong problem solving and programming skills in Python and/or C++
Strong culture fit and good team player.
Demonstrated research publications in top conferences (e.g. NeurIPS, ICLR, ICML, CVPR, RSS, CoRL etc.)
Preferred
No preferred qualifications listed.