AI Infrastructure Architect
Core
Architect end-to-end AI infrastructure (compute, fabric, storage, facilities) for customer AI factories, ensuring design authority across the lifecycle.
Role type
Senior IC AI Infrastructure Architect
Builds
AI factory deliverables (compute clusters, GPU platforms, high-performance fabrics)
Domain
Data Center / HPC / AI Infrastructure
Deliverable
production ML models
Required skills
NVIDIA DGX/HGX/GB200/NVL platforms, InfiniBand/Spectrum-X/RoCE, NVLink/NVSwitch, rack-scale architecture, HLD/LLD design, capacity planning, liquid cooling, digital twin simulation
Preferred skills
Large-scale HPC/supercomputing architecture, OEM validated designs (Cisco/Dell/HPE/Lenovo/Supermicro), NVIDIA infrastructure certification
Technologies
NVIDIA Air, Omniverse DSX, BlueField, ConnectX, UFM, Mission Control, NVSwitch, Spectrum-X
Responsibilities
Lead discovery and convert workload requirements into executable architecture; Select and document NVIDIA reference architectures; Design compute, GPU interconnect, fabric, storage, and facilities planes; Define rack elevation, power density, cooling, and failure domains; Produce HLD, LLD, BOM, topology, and test strategies; Chair architecture reviews and resolve cross-domain decisions; Define expansion units, lifecycle strategy, and Day 2 operating model; Mentor engineers through build and production handoff
Seniority
Senior, hands-on IC with leadership