Senior Systems Engineer: Rust-Driven Test Framework for GPUs

CoreWeave; Bellevue (WA); On-site; USD 153,000 - 204,000; Full time

Benefits offered by this job

Medical, dental & vision

Life Insurance

Supplemental life insurance

Disability Insurance

Flexible Spending Account

Health Savings Account

Tuition Reimbursement

Employee Stock Purchase Plan

Mental Wellness

Family-forming support

Parental Leave

Childcare support

401(k) match

Flexible PTO

Catered lunch

Casual work environment

Job summary

CoreWeave is seeking a Senior Software Engineer on the Systems Engineering team to own and evolve our test framework — a Kubernetes-native, in-house system that qualifies host images and infrastructure on real hardware. You’ll harden the framework's core, broaden coverage into HPC and Slurm-on-Kubernetes, and keep CI fast and trustworthy at fleet scale, where a 0.1% failure rate is thousands of GPUs.

Day-to-day, you’ll build test harnesses and CI infrastructure, extend coverage into HPC and

Qualifications

• 3+ years of experience building test infrastructure, systems software, or platform tooling at scale.

Responsibilities

• Own and evolve the in-house test framework for host images and infrastructure on real hardware.
• Broaden coverage to HPC, Slurm-on-Kubernetes (SUNK), and deeper hardware validation.
• Keep CI fast and trustworthy at fleet scale with hermetic tests and reduced flakiness.
• Design and maintain reporting paths, dashboards, and structured test results.
• Collaborate with firmware, kernel, imaging, and HPC teams to embed testing into release processes.

Skills

Test infrastructure

Python

Rust

Go/C/C++

Kubernetes

Linux

Tools

Argo Workflows

Job description

CoreWeave is seeking a Senior Software Engineer on the Systems Engineering team to own and evolve our test framework — a Kubernetes-native, in-house system that qualifies host images and infrastructure on real hardware. You’ll harden the framework's core, broaden coverage into HPC and Slurm-on-Kubernetes, and keep CI fast and trustworthy at fleet scale, where a 0.1% failure rate is thousands of GPUs.

Day-to-day, you’ll build test harnesses and CI infrastructure, extend coverage into HPC and