Senior Research Engineer, Foundation Model Training Infrastructure

NVIDIA•Santa Clara, CA

7d•$224,000 - $356,500

About The Position

NVIDIA is seeking a senior or principal engineer specializing in infrastructure for large-scale foundation model training to join the Generalist Embodied Agent Research (GEAR) group. The team is leading Project GR00T, NVIDIA’s initiative to build foundation models and full-stack technology for humanoid robots. The role supports research on multimodal foundation models, large-scale robot learning, embodied AI, and physics simulation, with contributions affecting research projects and product roadmaps. The engineer will design and maintain distributed training systems for multimodal robotics foundation models, optimize GPU and cluster utilization, implement scalable data-loading and preprocessing systems for video, text, and sensor data, and build monitoring and debugging tools for reliable training workflows on large GPU clusters. The role also collaborates with researchers to integrate new model architectures into scalable training pipelines.

Requirements

Bachelor's degree in Computer Science, Robotics, Engineering, or a related field.
10+ years of full-time industry experience in large-scale MLOps and AI infrastructure.
Proven experience designing and optimizing distributed training systems with frameworks such as PyTorch, JAX, or TensorFlow.
Deep understanding of GPU acceleration, CUDA programming, and cluster management tools such as Kubernetes.
Strong programming skills in Python and a high-performance language such as C++.
Strong experience with large-scale GPU clusters, HPC environments, and job scheduling or orchestration tools such as SLURM or Kubernetes.

Nice To Haves

Master’s or PhD degree in Computer Science, Robotics, Engineering, or a related field.
Demonstrated Tech Lead experience coordinating a team of engineers and driving projects from conception to deployment.
Strong experience building large-scale LLM and multimodal LLM training infrastructure.
Contributions to popular open-source AI frameworks or research publications at top-tier AI conferences such as NeurIPS, ICRA, ICLR, or CoRL.

Responsibilities

Design and maintain large-scale distributed training systems for multimodal foundation models for robotics.
Optimize GPU and cluster utilization for efficient model training and fine-tuning on massive datasets.
Implement scalable data loaders and preprocessors for multimodal datasets, including videos, text, and sensor data.
Develop monitoring and debugging tools to ensure the reliability and performance of training workflows on large GPU clusters.
Collaborate with researchers to integrate cutting-edge model architectures into scalable training pipelines.