Senior GPU AI Data Center Architect – Confidential Compute

Simulation 95; Menlo Park (CA); Hybrid; USD 165,000 - 248,000; Full time

Job summary

Simulation95 seeks a seasoned engineer to lead sourcing, design, and deployment of a private AI data center in a hybrid, contract role. You will manage hardware procurement, facility planning, security design, and orchestration integration for GPU infrastructure.

Expect growth from 24 to 48 to 104 NVIDIA GPUs, DGX-style nodes, InfiniBand networking, high-performance storage, and confidential computing with attestation and key management.

Qualifications

• Experience building and operating multi-node NVIDIA GPU clusters.
• Practical experience with InfiniBand, RDMA, NCCL, and high-performance storage.
• Experience with hardware-backed confidential computing and key management.
• Ability to assess data protection at rest, in transit, and in use.
• Strong Linux, virtualization, containerization, and automation skills.
• Experience sourcing enterprise hardware and coordinating vendors.

Responsibilities

• Develop BOMs, evaluate quotes, verify compatibility, and manage lead times.
• Assess electrical capacity, rack density, cooling, and connectivity with specialists.
• Design initial cluster and expansion plan for bandwidth, storage, and management.
• Deploy and tune InfiniBand, RDMA, and shared storage for workloads.
• Define threat models and validate isolation and attestation across system.
• Design encryption and access controls for datasets, model weights, and backups.
• Deploy Linux, NVIDIA software, containers, scheduling, and orchestration.
• Test performance, security, recovery, and reliability; document procedures.

Skills

GPU clusters

Vendor coordination

Security architecture

Troubleshooting

Linux administration

Technical communication

Tools

InfiniBand

RDMA

NCCL

Kubernetes

Slurm

DGX/NVIDIA software

Job description

Simulation95 seeks a seasoned engineer to lead sourcing, design, and deployment of a private AI data center in a hybrid, contract role. You will manage hardware procurement, facility planning, security design, and orchestration integration for GPU infrastructure.

Expect growth from 24 to 48 to 104 NVIDIA GPUs, DGX-style nodes, InfiniBand networking, high-performance storage, and confidential computing with attestation and key management.