AI Infrastructure Engineer

Acceler8 Talent; San Francisco, Northern (CA, KY); On-site; USD 180,000 - 300,000; Full time

Benefits offered by this job

Cash bonus

Founding engineer equity

Benefits

Job summary

Acceler8 Talent partners with an early-stage AI infrastructure company to hire a Founding AI Infrastructure Engineer in San Francisco. You’ll build and operate production GPU clusters across Kubernetes and Slurm, shape the hardware/software stack, and own performance, reliability and incident response for active workloads.

The role offers founding-engineer equity, competitive salary and benefits, with direct collaboration with the founders to deliver scalable AI compute and power-aware

Qualifications

• Hands-on experience building and operating Kubernetes or Slurm clusters from scratch.
• Strong knowledge of GPU infrastructure, distributed storage and high-performance networking.
• Experience owning production systems and resolving complex infrastructure issues.
• Experience with inference serving, distributed training or power-aware computing is highly valuable.

Responsibilities

• Building and operating production GPU clusters across Kubernetes and Slurm.
• Developing GPU orchestration, scheduling and lifecycle automation.
• Designing distributed storage, NVMe and high-bandwidth networking systems.
• Owning observability, reliability, workload performance and incident response.
• Working directly with customers to translate workload needs into infrastructure.

Tools

Kubernetes

Slurm

GPU Infrastructure

Distributed Storage

High-Performance Networking

Observability

Power-Aware Computing

Job description

Founding AI Infrastructure Engineer

Founding Infrastructure Engineer – GPU Neocloud & AI Infrastructure

San Francisco, CA – On-site

I’m working with an early-stage infrastructure company building the first flex-power neocloud.

The company is building the hardware and software stack to keep GPU workloads reliable while managing compute around power availability. This could unlock approximately 70GW of existing capacity and bring AI compute online in months rather than waiting five or more years for new grid infrastructure.

This is a true founding engineering opportunity, working directly with the founders to build and operate production infrastructure for active customer workloads.

You’ll work on:

• Building and operating production GPU clusters across Kubernetes and Slurm
• Developing GPU orchestration, scheduling and lifecycle automation
• Designing distributed storage, NVMe and high-bandwidth networking systems
• Owning observability, reliability, workload performance and incident response
• Working directly with customers to translate workload needs into infrastructure

Requirements:

• Hands-on experience building and operating Kubernetes or Slurm clusters from scratch
• Strong knowledge of GPU infrastructure, distributed storage and high-performance networking
• Experience owning production systems and resolving complex infrastructure issues
• Experience with inference serving, distributed training or power-aware computing is highly valuable

The company was founded by experienced second-time (successful!) entrepreneurs from AI infrastructure, software, power, real estate and data centres.

Despite being at the founding-team stage, the business already has approximately $5M in annual contract value under contract, a further $40M in pipeline and ongoing partnership discussions with NVIDIA and AMD.

Package: Competitive salary, cash bonus, benefits and meaningful founding-engineer equity.