Principal AI Researcher — World Foundation Models

NVIDIA Corporation; California (MO); Hybrid; USD 184,000 - 357,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is building the next generation of AI systems that perceive, reason about, and generate dynamic worlds. This role focuses on human appearance, motion, and action understanding, developing and validating model improvements that translate into production-grade checkpoints and recipes.

You will work with data, platform, and product engineering to ensure real-world performance. The ideal candidate has a PhD or equivalent experience, 8+ years in applied research, and 3+ years shaping

Qualifications

• PhD in Computer Science, Graphics, Computer Engineering, or closely related field, or equivalent experience.
• 8+ years of applied research and/or industry experience in vision, graphics, or adjacent ML domains.
• 3+ years designing, training, and evaluating generative models for image/video, with strong fundamentals in modern deep learning.
• Hands‑on experience improving generative models focusing on perceptual quality and temporal stability, esp. for generating humans.
• Advanced proficiency in Python, PyTorch, C++, and CUDA with strong research-engineering practices (reproducibility, testing, profiling, experiment tracking).
• Experience training and debugging large models in multi-GPU and/or multi-node environments and distributed training workflows.
• Practical knowledge of inference/runtime bottlenecks and optimization techniques.
• Strong eye for quality and ability to diagnose visual artifacts using perceptual metrics or learned evaluators.

Responsibilities

• Research, implement, and validate model architecture and algorithm changes to improve video generation fidelity with a focus on human-centric quality.
• Prototype improvements across spatial multimodal modeling and diffusion-based video generation to improve controllability and long-horizon consistency.
• Improve training and inference efficiency through architectural and post-training techniques (compute/memory optimizations, distillation, pruning, compression).
• Define model training objectives that improve sim-to-real and real-to-sim generalization for human motion and interactions.
• Develop detailed benchmarks for evaluating world foundation models in video, simulation, and physical environments.
• Translate research results into robust implementations like training code, checkpoints, model integrations, and demos.

Skills

Python

PyTorch

C++

CUDA

Generative models

Video generation

Distributed training

Research engineering

Education

PhD in Computer Science or related field

Tools

Multi-GPU training

Benchmarks

Job description

NVIDIA is building the next generation of AI systems that perceive, reason about, and generate dynamic worlds. This role focuses on human appearance, motion, and action understanding, developing and validating model improvements that translate into production-grade checkpoints and recipes.

You will work with data, platform, and product engineering to ensure real-world performance. The ideal candidate has a PhD or equivalent experience, 8+ years in applied research, and 3+ years shaping