GPU Infra Solutions Architect for Large-Scale AI Clusters

Prime Intellect; San Francisco (CA); On-site; USD 150,000 - 300,000; Full time

Job summary

Prime Intellect in San Francisco seeks a Solutions Architect for GPU Infrastructure who will transform client requirements into robust systems capable of training advanced AI models. Responsibilities include designing GPU cluster architectures, deploying orchestration systems, and supporting clients in optimizing their infrastructure.

The ideal candidate will have extensive experience with SLURM, Kubernetes, and NVIDIA architectures. A competitive cash compensation range of $150-300k plus equity incentives is provided.

Qualifications

• 3+ years hands-on experience managing GPU clusters in HPC environments.
• Deep expertise with orchestration tools SLURM and Kubernetes.
• Strong understanding of NVIDIA architectures and CUDA ecosystems.

Responsibilities

• Design GPU cluster architectures to meet client workload requirements.
• Deploy orchestration systems for distributed workloads.
• Act as technical escalation point for customer issues in infrastructure.

Skills

GPU clusters and HPC environments

SLURM

Kubernetes

InfiniBand configuration

NVIDIA GPU architecture

Python

Bash

Tools

Ansible

Terraform

Docker

Containerd

Job description

Prime Intellect in San Francisco seeks a Solutions Architect for GPU Infrastructure who will transform client requirements into robust systems capable of training advanced AI models. Responsibilities include designing GPU cluster architectures, deploying orchestration systems, and supporting clients in optimizing their infrastructure.

The ideal candidate will have extensive experience with SLURM, Kubernetes, and NVIDIA architectures. A competitive cash compensation range of $150-300k plus equity incentives is provided.