Senior Product Architect, HPC and Networking

Details

NVIDIA is the engine of modern Artificial Intelligence, Advanced Networking, and High Performance Computing (HPC). We’re on a mission to redefine how data moves, connects, and accelerates workloads at scale. As an HPC Networking Product Architect, you will design and shape architectures that connect the world’s most powerful AI clusters and drive networking-centric products and solutions from conception through lifecycle.

Responsibilities

• Architect datacenter-scale networking infrastructure leveraging SpectrumX Ethernet, InfiniBand, NVLink Switches, and other interconnect technologies.
• Design network and hardware-focused architectures that enable high-bandwidth, low-latency fabrics for NVIDIA GPUs, DPUs, compute and storage servers.
• Stay current on networking trends including software defined networking (SDN), programmable fabrics, congestion control, and datacenter interconnects.
• Define detailed product architectures covering performance, scalability, interoperability, and datacenter deployment requirements.
• Lead prototyping, benchmarking, and validation of networking-focused designs at scale; diagnose bottlenecks and drive performance innovations.
• Collaborate with internal hardware, software, security, and systems teams to ensure integration of networking, processing, and storage systems.
• Partner with customers and ecosystem partners to co-innovate, deliver proof-of-concepts (POCs) and MVPs, and contribute to developer and customer enablement.
• Create technical collateral including whitepapers, reference architectures, and developer documentation to guide adoption and best practices.

Requirements

• 12+ years of experience architecting datacenter-scale HPC or networking infrastructure in roles such as Principal Architect, Solutions Architect, or Principal Engineer.
• Bachelor’s or Master’s degree in Computer Science or a related field (or equivalent experience).
• Deep expertise in networking technologies including high-performance Ethernet (SpectrumX), InfiniBand, and NVLink Switches/Fabrics.
• Understanding of DPU (or similar SmartNIC) architectures and their role in SDN, security enforcement, and storage acceleration.
• Strong background in datacenter-scale network design: topologies, routing, congestion management, and fabric scalability.
• Experience with benchmarking tools and performance analysis for large-scale HPC/AI networking deployments.
• Knowledge of network isolation using VLAN, VxLAN, and EVPN overlays through SDN.
• Familiarity with Kubernetes networking, including CNI and Multus, and workload isolation at scale.
• Excellent communication skills, with the ability to translate technical networking concepts into business and product insights.

Ways to Stand Out

• Hands-on experience designing and deploying large-scale multi-tenant enterprise networking fabrics for AI/HPC workloads.
• Contributions to open standards or open-source projects in software-defined networking and programmable fabrics.
• Experience integrating Kubernetes, Multus, and advanced CNI plugins in high-performance environments.
• Published work, patents, or advanced certifications in networking or HPC systems.

Compensation & Benefits

• Base salary ranges by level:
  • Level 5: 224,000 USD - 356,500 USD
  • Level 6: 272,000 USD - 425,500 USD
• You will also be eligible for equity and benefits.

Additional Information

• Location: Santa Clara, CA (US)
• Office policy: Hybrid (#LI-Hybrid)
• Applications accepted at least until October 5, 2025.
• NVIDIA is an equal opportunity employer committed to diversity and inclusion.