Description
We are looking for an experienced Network Solutions Architect Engineer to help bring our next-generation AI networking platforms into production at customer data centers.
As part of the NVIDIA SA organization, you will be driving deployment of our end-to-end technology solutions integration at some of NVIDIA's most strategic technology customers, as well as offering recommendations to business and engineering teams on our product roadmap.
Your primary responsibilities will include:
Partnering with AI-native / consumer internet customers on large data center GPU and networking deployments. Guiding architecture decisions across network, compute, and storage, including fabric design. Supporting on-site bring-up of server, network, and cluster infrastructure in customer data centers.
Demonstrating expertise on advanced GPU and network systems (Spectrum-X, BlueField DPU, InfiniBand/RoCE, etc.) for key accounts. Running regular technical account reviews covering roadmap alignment, cluster issues, feature discussions, and new technology introductions. Capturing customer-specific requirements and translating them into concrete feedback for product, architecture, and engineering teams.
Analyzing and debugging configuration and performance issues in RoCE and InfiniBand environments. Working across NICs, switches, Linux, and system software to deliver performant, reliable AI clusters.
Identifying and shaping new project opportunities for NVIDIA GPUs, networking, and software in AI and data center use cases. Collaborating closely with Systems Engineering, Product Management, and Sales to align solutions with customer outcomes. Building targeted POCs that showcase the value of NVIDIA’s networking stack (e.g., Spectrum-X fabrics, BlueField DPUs) in real customer environments.
To be successful in this role, you will need:
A BS/MS/PhD in Electrical/Computer Engineering, Computer Science, or other Engineering fields or equivalent experience.
6+ years of hands-on network engineering experience in data center or cloud environments.
Proven, expert-level troubleshooting of data center networks (packet-level, control plane, and fabric behavior).
Deep protocol knowledge of BGP, OSPF, and L2/L3 switching in large-scale data center or cloud networks (ECMP, Clos/leaf–spine). Experience with high-density switching at cloud or hyperscale is strongly preferred. Experience with InfiniBand or RoCE is a major plus.
Solid understanding of CPU/GPU server architecture, NICs, Linux, system software, and kernel drivers
Strong time management and ability to context-switch across multiple customers and projects.
Excellent written and verbal communication, including clear design docs, customer presentations, and root-cause summaries.
Advanced certifications: CCIE, JNCIE, or equivalent expert-level certifications.
Automation & tooling: Experience in Python, Bash, or C/C++ for automating network workflows, validation, and debug.
NVIDIA platform experience: Hands-on work with NVIDIA GPUs, NICs, DPUs, or ARM-based CPU platforms.
Customer-facing background: Pre-sales, post-sales, field engineering, or consulting experience with external enterprise or cloud customers.
Large-scale deployments: Direct experience bringing up and operating large clusters or supercomputing environments.
Virtualization / cloud: Familiarity with virtualization, containers, and cloud networking concepts.
We make extensive use of conferencing tools, but occasional (20%) travel is required for on-site visit to customers and industry events. We are open to remote work location and look forward to have you join our team!