Founding HPC Engineer - GPU Compute & Infra (SF)

Hamilton Barnes ?; San Francisco (CA); On-site; USD 180,000 - 250,000; Full time

Benefits offered by this job

Relocation provided

Founding-level ownership

Job summary

Neocloud seeks a founding member to build and operate GPU-focused compute infrastructure in San Francisco. You will directly influence the platform that powers AI workloads, with ownership and visibility from the founders.

This onsite role relocates to San Francisco as needed. You will stand up production Kubernetes/Slurm clusters with custom GPU orchestration, design storage and networking for high-scale workloads, and own incident response end-to-end.

Qualifications

• Operate production Kubernetes and/or Slurm clusters with custom GPU orchestration and lifecycle automation.
• Deep hands-on experience with distributed object storage and NVMe storage clusters for large-scale AI workloads.
• Design and operate scalable inference platforms for low-latency, high-throughput GPU acceleration.
• Own end-to-end reliability and incident response; document runbooks and processes for new hires.

Responsibilities

• Stand up and operate Kubernetes/Slurm clusters with custom GPU orchestration.
• Design and implement storage architecture (object storage, NVMe clusters, high-bandwidth networking) for real customer workloads.
• Translate workload requirements into infrastructure decisions with customers.
• Own cluster reliability and production incident response end-to-end; document runbooks for future hires.

Skills

Kubernetes

Slurm

GPU orchestration

Cluster reliability

Distributed object storage

NVMe storage

High-bandwidth networking

AI workloads optimization

Incident response

Job description

Neocloud seeks a founding member to build and operate GPU-focused compute infrastructure in San Francisco. You will directly influence the platform that powers AI workloads, with ownership and visibility from the founders.

This onsite role relocates to San Francisco as needed. You will stand up production Kubernetes/Slurm clusters with custom GPU orchestration, design storage and networking for high-scale workloads, and own incident response end-to-end.