Senior Research Scientist

Job Summary

Focused on advancing generative video models through physics simulation and reinforcement learning, the full-time Senior Research Scientist will enhance motion reward systems and improve model dynamics while working remotely.

Key responsibilities

• Improving physics-based motion rewards for generative video models
• Researching and pushing the boundaries of state-of-the-art video generation models
• Extending model capabilities to include human-object and human-scene interactions using 3D mesh representations

Required qualifications

• Master's degree in robotics, graphics, mechanical engineering, computer science, or equivalent experience
• 5+ years of experience in reward design and post-training of large generative models
• Proficiency in rigid-body/contact simulation technologies (e.g., MuJoCo, Isaac, Bullet)
• Expertise in Python/PyTorch with a focus on optimizing compute-heavy pipelines
• Experience with 3D human pose and motion representations (e.g., SMPL-family)