Principal Solutions Architect (Req #1341)
eplusIrvine, CA
Principal Solutions Architect (Req #1341)
eplusIrvine, CAManager L7
Computer Systems Engineers/ArchitectsNetwork and Computer Systems AdministratorsComputer Network Architects
Computer Systems Design ServicesComputing Infrastructure Providers, Data Processing, Web Hosting, and Related ServicesComputer Facilities Management Services
Overview:
As a Solutions Architect, you will design end-to-end NVIDIA AI Factory-aligned infrastructure for production AI workloads. You collaborate with sales, product, and engineering to translate AI/ML requirements into scalable, multi-domain solutions spanning colocation, GPU compute, networking, storage, and software. You influence proposals, bill-of-materials, and architecture outcomes, shaping how enterprises deploy frontier models and large-scale inference. This role blends technical depth with client-facing leadership to accelerate transformational AI programs. Compensation / Benefits medical benefits 401(k) eligibility employee stock purchase program paid time off (vacation, sick leave, personal leave) flexible compensation components (bonuses, commissions)
Responsibilities:
Lead discovery workshops to capture AI/ML workload requirements Architect full-stack AI Factory solutions aligned to NVIDIA reference architectures Create BOMs, rack elevations, network topologies, and power/cooling budgets Define GPU cluster architectures using DGX/HGX/MGX systems and Blackwell configurations Design RTX PRO 6000 Blackwell Server Edition deployments for enterprise workloads Perform workload sizing and TCO/ROI modeling for training, finetuning, and inference Specify colocation requirements and coordinate with data center operators Architect multi-node GPU clusters optimized for LLM pre-training, RLHF, and fine-tuning Design high-performance GPU fabric networks using InfiniBand and Spectrum-X Configure RDMA, NCCL, and RoCEv2; integrate BlueField-3 DPUs Design parallel storage architectures with VAST, Hammerspace, and Pure Storage Configure data pipelines from cloud object stores to local storage Deploy and optimize NVIDIA AI Enterprise software stack and orchestration tools Implement Kubernetes-based AI platforms with Run:ai for multi-tenant workloads Setup SLURM for HPC-style scheduling and preemption policies Lead POC/pilot deployments and executive-level briefings Collaborate with procurement, logistics, and deployment teams Contribute to knowledge bases and repeatable AI Factory playbooks
Key requirements:
8+ years in solutions architecture, systems engineering, or technical pre-sales 4+ years focused on GPU infrastructure or HPC environments Proven record deploying NVIDIA DGX or HGX-based GPU clusters Deep understanding of distributed DL concepts (tensor, pipeline, data parallelism) Hands-on experience with InfiniBand or high-speed Ethernet fabric, RDMA, and NCCL/MPI tuning Experience sizing and deploying parallel storage (VAST, Hammerspace, Lustre/WEKA/GPFS) Strong knowledge of Kubernetes, GPU Operator, and a GPU workload scheduler (Run:ai or SLURM) Linux administration, CUDA development environment setup, and firmware management Ability to produce architecture diagrams and BOM-level documentation Excellent communication for technical and exec audiences Executive presence Customer obsession Collaborative leadership DGX B200/B300, DGX H100/H200, HGX B200/B300, HGX H100/H200, MGX, GB300 NVL72/GB200 NVL72 RTX PRO 6000 Blackwell Server Edition NVLink Switch System
Matching similar jobs
Level
Manager L7
Location
Irvine, CA
Occupation
Computer Systems Engineers/Architects
Industry
Computer Systems Design Services
Posted
today