Senior Research Scientist — Physics-based Generative AI

Crystal Equation Corporation; California (MO); Hybrid; USD 109,000 - 116,000; Full time

Benefits offered by this job

Paid time off

Medical insurance

Dental insurance

Vision insurance

Life insurance

401K

Parental leave

And more

Job summary

Crystal Equation Corporation seeks a Research Scientist V to advance physics-based rewards for post-training generative video models using reinforcement learning. You will own the reward signal, drive performance across three workstreams, and extend capabilities to human-object and human-scene interaction with 3D mesh representations.

The role requires expertise in rigid-body simulation, 3D pose representations, and Python/PyTorch optimization, with a PhD/MS in a relevant field or equivalent

Qualifications

• MS/PhD in robotics, graphics, mechanical engineering or CS, or equivalent industry experience

Responsibilities

• Owns a physics-based motion reward used to post-train generative video models via reinforcement learning.
• Three workstreams: improving the reward signal terms; making the reward cheap enough to run inside the loop at scale; extending it from single-body plausibility to human-object and human-scene interaction using 3D mesh representations.

Skills

Rigid-body simulation

3D pose & motion

Python/PyTorch

Reward design/post-training

Numerical methods

Education

MS/PhD in robotics/graphics/mechanical engineering or CS

Tools

MuJoCo

Isaac

Bullet

SMPL-family

Job description

Crystal Equation Corporation seeks a Research Scientist V to advance physics-based rewards for post-training generative video models using reinforcement learning. You will own the reward signal, drive performance across three workstreams, and extend capabilities to human-object and human-scene interaction with 3D mesh representations.

The role requires expertise in rigid-body simulation, 3D pose representations, and Python/PyTorch optimization, with a PhD/MS in a relevant field or equivalent