Senior DL Infrastructure Engineer – Scalable GPU Training

NVIDIA Corporation; California (MO); Hybrid; USD 224,000 - 431,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA Corporation seeks a Deep Learning Software Infrastructure Engineer to advance the Autonomous Vehicles initiative. You will build and scale training libraries and infrastructure for multi-thousand GPU clusters, collaborating with researchers and platform teams to improve training speed and reliability.

You should have a PhD/MS/BS in CS/EE and 12+ years of distributed systems experience, with PyTorch and Python proficiency, plus expertise in datacenter networking and large-scale data

Qualifications

• BS/MS/PhD in Computer Science, Electrical/Computer Engineering, or a related field, or equivalent experience.
• 12+ years of professional experience building and scaling high-performance distributed systems.
• Extensive knowledge in deep learning frameworks (PyTorch preferred) and large-scale training.
• Strong systems background: datacenter networking (RoCE, IB), parallel filesystems (Lustre), storage systems, schedulers (Slurm, Kubernetes).
• Proficiency in Python with experience writing production-grade libraries and orchestration tools.
• Ability to collaborate with ML researchers, infra engineers, and product leads to translate requirements into robust systems.

Responsibilities

• Crafting, scaling, and hardening deep learning infrastructure libraries for training on multi-thousand GPU clusters.
• Improving efficiency across the training stack (data loaders, distributed training, scheduling, monitoring).
• Building robust training pipelines for massive video datasets and rapid experimentation.
• Collaborating with researchers and platform teams to improve training availability and stability.
• Owning core infrastructure components such as orchestration libraries and fault-resilient training systems.
• Partnering with leadership to ensure infrastructure scales with GPU capacity and dataset size.

Skills

Python

Distributed systems

PyTorch

Kubernetes

NCCL

Slurm

Data infrastructure

Education

BS/MS/PhD in CS/EE

Tools

Kubernetes

Slurm

Lustre

RoCE

IB

Job description

NVIDIA Corporation seeks a Deep Learning Software Infrastructure Engineer to advance the Autonomous Vehicles initiative. You will build and scale training libraries and infrastructure for multi-thousand GPU clusters, collaborating with researchers and platform teams to improve training speed and reliability.

You should have a PhD/MS/BS in CS/EE and 12+ years of distributed systems experience, with PyTorch and Python proficiency, plus expertise in datacenter networking and large-scale data