AI Infrastructure Domain Architect
J ThejuUnited StatesSENIORPosted 9h ago
NVIDIA GPU infrastructureHPCaccelerated computingNVIDIA DGX/HGX platformsDell GPU server platformsHPE GPU server platformsNVIDIA DGX BasePODNVIDIA SuperPODNVIDIA Enterprise Reference ArchitecturesNVIDIA InfiniBandSpectrum-X EthernetRDMARoCEAI storage platformsDDNWEKAVASTPure StorageNVIDIA CUDADGX OSGPU OperatorNGCBase Command ManagerSlurmKubernetesdata center infrastructurerack power densityliquid/air coolingthermal requirementsHLD/LLDBOMnetwork designsrack layoutsdeployment documentation
About the Role
We are looking for an experienced AI Infrastructure Domain Architect with strong expertise in NVIDIA GPU infrastructure, HPC, and accelerated computing. The candidate will design and deliver enterprise-scale AI infrastructure solutions, including GPU compute, networking, storage, and data center infrastructure.
Key Responsibilities
• Design and deliver enterprise-scale AI infrastructure solutions
• Design GPU compute, networking, storage, and data center infrastructure
• Create HLD/LLD, BOM, network designs, rack layouts, and deployment documentation
• Present technical architecture and solutions to customers
Requirements
• 8–12 years in enterprise infrastructure architecture
• Minimum 3 years of GPU, HPC, or accelerated-compute infrastructure experience
• Strong hands-on experience with NVIDIA DGX/HGX platforms (B200, B300, H200, GB200, etc.)
• Experience with Dell or HPE GPU server platforms using NVIDIA GPUs
• Knowledge of NVIDIA DGX BasePOD, SuperPOD, and Enterprise Reference Architectures
• Strong expertise in NVIDIA InfiniBand, Spectrum-X Ethernet, RDMA, and RoCE
• Experience with AI storage platforms such as DDN, WEKA, VAST, or Pure Storage
• Knowledge of NVIDIA CUDA, DGX OS, GPU Operator, NGC, and Base Command Manager
• Working knowledge of Slurm and Kubernetes
• Strong understanding of data center infrastructure, rack power density, liquid/air cooling, and thermal requirements
• Must have delivered at least one multi-node GPU cluster from design through acceptance
• Strong customer-facing architecture and technical presentation skills
• Must be willing to travel as required
• Strong on-premises NVIDIA GPU infrastructure and data center experience is mandatory
• Candidates with only cloud-based GPU experience, AI/ML development, or MLOps experience without on-premises NVIDIA GPU cluster deployment will not be considered