Sr. Site Reliability Engineer, AI Infrastructure (Starshield)
SpaceX · Redmond, WA
About this role
**SpaceX — Sr. Site Reliability Engineer, AI Infrastructure (Starshield)** SpaceX is building the technologies to enable human life on Mars. Starshield is the world’s largest US government satellite constellation, providing immediate access to critical intelligence and national security data for the US government anywhere on the globe. As a Sr. Site Reliability Engineer focused on Starshield’s software and GPU infrastructure, you will design, operate, and scale infrastructure supporting critical national security missions. This role spans Site Reliability Engineering, Developer Operations, and GPU platforms. **Responsibilities** - Manage GPU/CPU infrastructure deployments to Top Secret data centers - Manage and support GPU-as-a-service for external customers on bare metal and virtualized platforms - Design, validate, and productize solutions for AI clusters (100k+ GPU scale) - Develop automation to deploy and manage on-premise Kubernetes/AI clusters and operating systems - Deploy and manage core infrastructure such as databases, monitoring, and distributed storage - Collaborate with AI engineers to build scalable, operable, and maintainable products - Own services across the full lifecycle: inception/design → deployment → operation → refinement - Implement monitoring and alerting to support high availability - Identify improvement opportunities and create innovative solutions to increase system availability - Mentor and train junior engineers - Lead the team to technical excellence **Basic Qualifications** - Bachelor’s degree in CS/IS/IT or an engineering discipline **and** 5+ years of experience with Linux; **or** 7+ years of experience in software, DevOps, or SRE in lieu of a degree - 5+ years of experience with Kubernetes - 5+ years of experience managing Linux operating systems *Note: The provided description appears truncated after the Basic Qualifications section.*
Listing freshness
CronJobs last confirmed this listing 1h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.