SRE for High-Performance Compute Platform

Spacex; Hawthorne (CA); Hybrid; USD 125,000 - 195,000; Full time

Benefits offered by this job

Stock options

401(k) retirement plan

Paid vacation & holidays

Job summary

SpaceX in Hawthorne, CA is seeking a Site Reliability Engineer for High Performance Computing to own the Linux, infrastructure as code, storage, and user-facing services across the HPC platform.

You will reduce toil through automation, build cluster observability, and collaborate with HPC systems engineers and engineers from across SpaceX to deliver reliable, scalable compute and storage for mission-critical workloads.

Qualifications

• Bachelor's degree or 2+ years production infra experience
• 2+ years Linux in production
• 2+ years operating production infrastructure, including monitoring, debugging, and repairs

Responsibilities

• Participate in on-call rotation
• Manage node lifecycle with infrastructure as code: OS images, firmware, configuration, kernel and driver stack
• Build observability for HPC admins and end users: cluster, node, and storage health signals
• Reduce toil with automation and write software to automate repetitive tasks
• Sustainably manage compute and storage resources
• Lead capacity planning with users across SpaceX to forecast tomorrow's compute/storage needs
• Collaborate with HPC engineers and other disciplines on operable, maintainable infrastructure

Skills

SRE/DevOps experience

Linux in production

Infrastructure as code

Monitoring & alerting

Python or scripting

Containers (Docker, Podman)

Kubernetes on-prem

Distributed storage

Education

Bachelor's degree or equivalent in CS/Engineering/Math/Science

Tools

Prometheus

Grafana

Nagios

Ansible

Terraform

Puppet

Docker

Kubernetes

Slurm

Job description

SpaceX in Hawthorne, CA is seeking a Site Reliability Engineer for High Performance Computing to own the Linux, infrastructure as code, storage, and user-facing services across the HPC platform.

You will reduce toil through automation, build cluster observability, and collaborate with HPC systems engineers and engineers from across SpaceX to deliver reliable, scalable compute and storage for mission-critical workloads.