Data Center GPU Solutions Architect Scale Deployments

NVIDIA AI; Seattle (WA); On-site; USD 152,000 - 288,000; Full time

Benefits offered by this job

Equity compensation

Benefits package

Job summary

NVIDIA seeks an Infrastructure Solutions Architect to lead deployment and bring-up of Data Center GPU and networking platforms. You will work across product, engineering, and cloud teams to accelerate adoption with large customers and hyperscalers.

Responsibilities include end-to-end execution, performance analysis, and defining unified performance metrics while steering cross-functional collaborations and executive communications.

Qualifications

• BS/MS/PhD in Electrical/Computer Engineering, CS, Physics, or equivalent experience.
• 4+ years experience in Solutions Architecture, Infrastructure Engineering, or similar technical roles.
• Hands-on bring-up and validation of large-scale NVIDIA GPU platforms, multi-GPU and multi-node architectures.
• Understanding of high-performance networking technologies and their role in distributed AI workloads.
• Familiarity with CUDA, NCCL, NVSwitch/NVLink, drivers and performance tuning.
• Proficiency with Linux systems tools for diagnosing issues and evaluating performance.
• Understanding server hardware architecture, PCIe topologies, NUMA, BIOS/UEFI, power/thermal envelopes.
• Knowledge of BMC/IPMI/Redfish for remote management during early bring-up.
• Strong Linux fundamentals across drivers, kernel subsystems, containers, and performance analysis.
• Ability to identify bottlenecks at cluster, node, accelerator, network or app layer.

Responsibilities

• Lead end-to-end execution for hyperscaler customers to bring NVIDIA GPU platforms to market at scale.
• Align with Product teams to define metrics and unified direction across teams.
• Influence across Product, Engineering, Sales, Operations, and CSP customers to unblock scale-up.
• Analyze deployment and performance data to identify health trends and risks.
• Solve technical problems involving GPUs, networking, drivers, containers, firmware, and distributed systems.
• Deliver executive-level status updates, risks, and decisions.
• Collaborate with Product and Engineering to improve platform design and workflows.

Skills

Infrastructure architecture

Hands-on GPU platform bring-up

Linux system fundamentals

High-performance networking

CUDA/NCCL/NVSwitch familiarity

Driver behavior and performance tuning

System software stacks (CUDA, NCCL, NV

Education

BS/MS/PhD in Electrical/Computer Engineering, CS, Physics or equivalent

Tools

dmesg

journalctl

lspci

numactl

ethtool

iostat

perf

nvidia-smi

top/htop

Job description

NVIDIA seeks an Infrastructure Solutions Architect to lead deployment and bring-up of Data Center GPU and networking platforms. You will work across product, engineering, and cloud teams to accelerate adoption with large customers and hyperscalers.

Responsibilities include end-to-end execution, performance analysis, and defining unified performance metrics while steering cross-functional collaborations and executive communications.