Senior AI Researcher, World Foundation Models

NVIDIA; Washington; On-site; USD 184,000 - 357,000; Full time

Benefits offered by this job

Equity program

Benefits package

Job summary

NVIDIA is hiring a senior applied researcher to advance world foundation models, focusing on human appearance, motion, and action understanding. You will validate architectures, prototype improvements, and translate results into production-grade code and demos.

The role requires PhD or equivalent, 8+ years in vision/graphics ML, and strong Python/PyTorch/C++/CUDA skills, with multi-GPU experience and a passion for perceptual quality.

Qualifications

• PhD in Computer Science, Graphics, or closely related field (or equivalent experience).
• 8+ years of applied research/industry experience in vision, graphics, or ML domains.
• 3+ years designing, training, evaluating generative models for image/video/audio.
• Hands-on experience improving generative models focused on perceptual quality and temporal stability.
• Advanced proficiency in Python, PyTorch, C++, and CUDA with reproducibility practices.
• Experience training and debugging large models in multi-GPU / multi-node environments.

Responsibilities

• Research, implement, and validate model architecture changes to improve video generation fidelity.
• Prototype improvements across spatial multimodal modeling and neural rendering-inspired representations.
• Improve training/inference efficiency via architectural and post-training optimizations.
• Define training objectives to improve generalization on human motion and interactions.
• Develop benchmarks for evaluating world foundation models in video and simulation contexts.
• Translate research results into production-grade code, checkpoints, and demos.

Skills

Python

PyTorch

C++

CUDA

Research

Experimentation

Education

PhD or equivalent

Tools

Distributed training

Profiling & debugging

Job description

NVIDIA is hiring a senior applied researcher to advance world foundation models, focusing on human appearance, motion, and action understanding. You will validate architectures, prototype improvements, and translate results into production-grade code and demos.

The role requires PhD or equivalent, 8+ years in vision/graphics ML, and strong Python/PyTorch/C++/CUDA skills, with multi-GPU experience and a passion for perceptual quality.