Senior AI Infrastructure Architect - GPU Scale (Remote)

NVIDIA Corporation; Santa Clara (CA); À distance; USD 184 000 - 357 000; Plein temps

Résumé du poste

NVIDIA Corporation in Santa Clara, CA, is seeking an AI Solutions Architect to design and scale GPU-based AI infrastructure, collaborating with leading consumer internet companies and frontier labs to accelerate foundation model workloads.

You will optimize large-scale AI clusters, profile workloads, and lead proof-of-concept studies, working with NVIDIA engineers, product, and sales to win design opportunities and deliver high-performance, observable systems.

Qualifications

• 6+ years of experience in AI infrastructure, systems engineering, HPC, networking, site reliability engineering, or related technical role.
• Deep understanding of Linux systems, distributed computing, GPU architectures, and large-scale AI clusters.
• Hands-on experience deploying/troubleshooting high-performance GPU networks using InfiniBand/RoCE/GPUDirect RDMA.
• Experience debugging NCCL and distributed performance, including topology and routing.
• Experience profiling AI workloads and bottlenecks across compute, networking, storage, and orchestration layers.
• Experience with Kubernetes and Slurm, containers, and production monitoring systems.
• Proficiency with Python and shell scripting for automation, benchmarking, and troubleshooting.

Responsabilités

• Collaborate with customers to maximize GPU utilization and throughput while improving reliability and reducing costs.
• Design and optimize large-scale AI clusters across GPU compute, networking, storage, scheduling, orchestration, and observability.
• Profile distributed training and inference workloads to identify bottlenecks across GPUs, CPUs, memory, network fabrics, and storage.
• Diagnose complex infrastructure issues spanning InfiniBand/RoCE, cloud interconnects, RDMA, NCCL, NVLink, and NVSwitch.
• Lead proof-of-concepts and performance studies; develop benchmarking tools, automation, and runbooks as needed.
• Partner with NVIDIA’s engineering, product, and sales teams to secure design wins based on customer requirements and field feedback.

Connaissances

GPU architectures

Linux systems

NCCL

InfiniBand networking

Kubernetes

Slurm

Python

Shell scripting

Performance profiling

Formation

BS/MS/PhD in CS/EE/Physics/Math or equivalent

Outils

NCCL tests

DCGM

Nsight Systems

Description du poste

NVIDIA Corporation in Santa Clara, CA, is seeking an AI Solutions Architect to design and scale GPU-based AI infrastructure, collaborating with leading consumer internet companies and frontier labs to accelerate foundation model workloads.

You will optimize large-scale AI clusters, profile workloads, and lead proof-of-concept studies, working with NVIDIA engineers, product, and sales to win design opportunities and deliver high-performance, observable systems.