Lead Network Architect – AI Infrastructure (GPU Clusters)
Nava
—
Jul 27, 2026
—
Jul 28, 2026
What this job asks for AI summary
This role leads the architectural design and hands-on deployment of high-performance networking fabrics for large-scale GPU clusters (thousands of GPUs) at an early-stage AI cloud startup. Day-to-day work spans InfiniBand and Ethernet fabric orchestration, BlueField DPU offloading strategies, RDMA/RoCE tuning, and automating network lifecycle management. It suits a deep networking specialist with proven experience operating CLOS/Leaf-Spine fabrics at GPU-cluster scale.
Senior level
We read this from the posting text with AI. Skim the description below before ruling yourself out.
How this req sits in the market our data
Most people in this occupation already list these. Still required — just not what gets you shortlisted.
What the occupation pays Median $101,310 (middle half $79,725–$129,425).
Estimated from BLS employment for this occupation and area, per-skill prevalence across our listing corpus, and published wage benchmarks — as of Jul 28, 2026. It is a model, not a headcount.
Why we read it this way (6)
Location is not stated anywhere in the posting — CBSA and state are left blank. The role may be remote or at an unstated office; remote=false is set conservatively since no remote claim is made.
The SOC classification is a close call: the role involves both architecting/operating physical network infrastructure (15-1244) and a meaningful Infrastructure-as-Code / software-defined automation component (15-1252). 15-1244 is chosen because the primary day-to-day work described is designing and operating the physical fabric, not shipping software.
NCCL, UCX, SHARP, and Infrastructure as Code appear in the role description narrative rather than under a distinct 'Required' heading; they are marked preferred accordingly.
Preferred qualifications (liquid-cooled rack networking, open-source/OCP contributions, TCO financial modeling) name no concrete named technologies and are omitted per extraction rules.
No compensation figures are provided; the $22M Series A mention is funding context, not a salary offer.
Ignored 2 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): NVIDIA Quantum-2 InfiniBand (NDR), CLOS/Leaf-Spine architecture.
Read the full posting
The employer publishes the full description on their own site — read it there ↗. Or sign in to read it here — it's free, and it also lets you track this application.