Tech Stack

Description

You will develop and scale state-of-the-art generative models for autonomous plan generation, leveraging diffusion architectures, flow-matching, and energy-based models. You'll collaborate across autonomy teams, deploying your models on the NuroDriver to solve top autonomy challenges.

Requirements

• PhD preferred or MSc with 2-3 years experience in generative models
• Research experience in diffusion models, flow matching, energy-based models for robotics
• Strong problem-solving and programming skills in Python and/or C++
• Demonstrated research publications in top conferences (NeurIPS, ICLR, ICML, CVPR, RSS, CoRL)
• Experience with vision-language-action models, video generation, or world foundation models is a bonus

Responsibilities

• Develop and scale state-of-the-art generative models for autonomous plan generation
• Build generative models with foundation models, including LLMs and world foundation models
• Optimize generative models using reinforcement learning for interactive reasoning
• Explore joint prediction and planning, self-play, and active learning
• Collaborate across autonomy teams and deploy models on NuroDriver