Software Engineer - Bare-metal GPU Infrastructure

Metric; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Metric, a GPU-as-a-Service start-up, is seeking a Software Engineer focusing on the bare-metal GPU platform. You will help build the foundation for delivering high-performance GPU resources as customer-ready infrastructure and influence automation, qualification, productionization, and ongoing operation of GPU hardware.

You will join as an early team member in the Bay Area, working onsite to design cloud-native provisioning, qualification, and lifecycle processes, and to establish scalable,

Qualifications

• Experience in bare-metal platform engineering or systems/SRE roles.
• Hands-on Linux, Kubernetes, GPUs, DPUs, and infrastructure automation experience.
• Experience with GPU diagnostics, NVIDIA networking/RDMA, or distributed storage is a plus.

Responsibilities

• Develop robust, cloud-native provisioning solutions for servers and GPUs.
• Design qualification, simulation, and evaluation processes for GPU, storage, and network infrastructure.
• Model and optimize fleet-wide lifecycle workflows, including monitoring, repair, and requalification of hardware.
• Establish best practices for scalable, resilient, and secure infrastructure.

Skills

Bare-metal platform engineering

SRE / infrastructure software

Cloud-native infrastructure design

Tools

Linux

Kubernetes

GPUs

DPUs

Infrastructure automation

Job description

Metric are partnered with an exciting GPU-as-a-Service start-up, currently in stealth mode.

As a Software Engineer focusing on the bare-metal GPU platform, you will play a key role in building the foundation for delivering high-performance GPU resources as customer-ready infrastructure. As an early team member, you will influence the automation, qualification, productionization, and ongoing operation of GPU hardware critical to modern AI.

Contract: Permanent

Onsite: Onsite in the Bay Area

The Role

• Develop robust, cloud-native solutions for fast and secure provisioning of servers and GPUs.
• Design and implement qualification, simulation, and evaluation processes for GPU, storage, and network infrastructure.
• Model and optimize fleet-wide lifecycle workflows, encompassing monitoring, repair, and requalification of hardware.
• Establish best practices to ensure scalable, resilient, and secure infrastructure.

Experience

• Experience working in bare-metal platform engineering, infrastructure software development or systems/SRE
• Hands-on experience with Linux, Kubernetes, GPUs, DPUs, and infrastructure automation.
• Experience in areas such as or related to GPU diagnostics, NVIDIA networking/RDMA, distributed storage architectures or Kubernetes orchestration