AI Networking Domain Architect

GDH

1h ago • Remote

Senior

Remote

208,000 - 270,400 USD/yr

  1. Select offer
  2. Add resume
  3. Analyze
  4. Improve
  5. Apply
  6. Auto-send · soon

Match your resume with similar open roles

Upload your resume to automatically match with active openings similar to this role.

PDF up to 10 MB

Your resume is private & secure

Potential salary for AI Networking Domain Architect role

Improve your resume now

$208,000 – $270,400/year

Quick Facts

  • Focus: High-performance AI networking fabrics (InfiniBand and Ethernet) across architecture, implementation, automation, telemetry, and performance engineering.
  • Role split: ~60% delivery of complex AI infrastructure; ~40% technical scoping, solution design, and pre-sales/commercial support.
  • Work model: Fully remote from an approved location.
  • Pay range: $100.00 – $130.00 hourly.

Description

This senior-level role is the technical authority for the physical and logical lifecycle of high-performance InfiniBand and Ethernet interconnect fabrics. You will combine hands-on architecture and implementation with customer consulting and pre-sales support for large-scale AI infrastructure deployments.

Responsibilities

  • Architect and deploy non-blocking fat-tree (Clos) topologies using NVIDIA Quantum-2 NDR, Quantum-3 XDR, and Spectrum-4 Ethernet switches.
  • Design rail-optimized fabrics that align GPU-to-GPU traffic with compute-node PCIe topology and optimize NCCL collective operations.
  • Configure adaptive routing, congestion control, QoS, RoCEv2, PFC, and ECN for lossless, high-performance networking.
  • Enable and tune SHARP in InfiniBand environments; architect SmartNIC/DPU offload strategies using NVIDIA BlueField-3 and DOCA.
  • Deploy NVIDIA UFM, NetQ, and PTP for fabric management, telemetry, and nanosecond-level clock synchronization.
  • Build Python and Ansible infrastructure-as-code to automate switch configuration, validate topology designs, and support ongoing operations.
  • Define physical cabling strategies (DAC, AOC, OSFP, QSFP112, breakout configurations) and validate signal integrity and optical link budgets for 400G and 800G.
  • Run acceptance testing with NCCL tests and monitor fabric health to identify slow receivers, link degradation, and performance bottlenecks.
  • Calculate workload bisectional bandwidth needs, produce level-of-effort estimates, and validate network bills of materials against hardware compatibility lists.
  • Lead client workshops and integrate AI back-end fabrics with front-end management networks, including BGP and EVPN handoffs.

Requirements

  • 10+ years of experience in networking, high-performance computing, or data center infrastructure engineering, including at least 7 years in a customer-facing architecture role.
  • Relevant education, training, certifications, or equivalent practical experience.
  • Deep knowledge of NVIDIA networking platforms: Quantum InfiniBand switch architecture, MLNX-OS, UFM, Spectrum-X Ethernet, NVIDIA Cumulus Linux, and SONiC.
  • Proficiency in RoCEv2, RDMA, PFC, ECN, NVIDIA BlueField DPUs, and DOCA offload use cases.
  • Experience with Python and Ansible for switch configuration management and infrastructure automation.
  • Ability to translate complex technical architectures into commercial outcomes for senior stakeholders.
  • Understanding of BGP, EVPN, and VXLAN for multi-tenant network isolation.
  • Knowledge of host-network integration (IPoIB, Netlink) and GPUDirect RDMA plus NCCL communication patterns.
  • Availability to support remote client engagements, technical workshops, infrastructure delivery, and pre-sales activities.

Benefits

  • Fully remote role from an approved location.
  • Publishing pay range: $100.00 – $130.00 hourly.

See similar jobs

Explore similar jobs and use your resume score to choose where to apply next

See more

Unlock Full Resume Report