Apply Edge Start your job search

Senior Platform Engineer — GPU Cloud Infrastructure

StratITech · San Francisco Bay Area

Apply & track with Apply Edge

Senior Platform Engineer — GPU Cloud Infrastructure

Location: San Francisco, CA — Onsite 5 days/week

Employment Type: Full-TimeCompensation: $210,000–$260,000 base + equityWork Authorization: Candidates must be currently authorized to work in the United States.

This position is not eligible for new or future employer-sponsored work authorization.Note: No C2C arrangements will be considered.Any attempt to use personal or household contact information for solicitation, candidate submission, or vendor outreach is strictly prohibited and will be reported to LinkedIn.About the RoleWhat would it be like to help build a cloud before all of the important infrastructure decisions have already been made?We are partnering with a highly technical, well-funded AI company launching a new GPU-focused neocloud designed to provide high-performance accelerated computing infrastructure.We are looking for a Senior Platform Engineer to help build the foundational systems behind this new GPU cloud. This is a true 0→1 infrastructure role where you will help determine what should be built, make architectural decisions, build the systems, and own them through production.This is not a traditional production-operations SRE role focused primarily on maintaining an established stack. The environment spans Kubernetes, Linux, networking, storage, Infrastructure as Code, distributed systems, observability, automation, and GPU-accelerated compute.What You'll DoDesign and build foundational infrastructure for a new GPU cloud platform.Own major infrastructure projects from architecture through implementation and production.Design and evolve Kubernetes-based infrastructure for distributed, compute-intensive workloads.Build platform software, automation, controllers, and internal tooling using Python and/or Go.Develop reusable infrastructure using Terraform / Infrastructure as Code.Build deployment, provisioning, orchestration, and lifecycle-management capabilities.Develop self-service infrastructure that makes complex compute resources easier to consume.Work across Linux, networking, storage, containers, Kubernetes, and distributed systems.Build observability and diagnostic capabilities for large distributed environments.Troubleshoot complex issues across application, OS, network, storage, container, and infrastructure boundaries.Help establish technical patterns and architecture while the platform is still being created.What We're Looking ForApproximately 8+ years of relevant engineering experience.Demonstrated experience building infrastructure or platforms, not primarily operating systems built by others.Proven ownership of meaningful projects from architecture through production.Strong Linux and systems fundamentals.Deep hands-on experience with production Kubernetes.Strong Terraform / Infrastructure as Code experience.Production programming or substantial infrastructure automation experience with Python, Go, or another systems-oriented language.Experience building or significantly extending CI/CD, deployment systems, developer infrastructure, or platform tooling.Strong understanding of distributed systems and cross-layer troubleshooting.Strong architectural judgment and ability to explain technical tradeoffs.Comfort working in a 0→1 environment with incomplete requirements and significant individual ownership.Desire to remain deeply hands-on.Especially Valuable ExperienceWe are particularly interested in engineers who have:Built a platform or infrastructure capability from scratch.Served as an early, founding, or first infrastructure/platform engineer.Built internal developer platforms or self-service infrastructure.Written Kubernetes operators, controllers, schedulers, or provisioning systems.Built distributed compute, storage, networking, or data infrastructure.Independently turned ambiguous technical objectives into production systems.Nice to HaveExperience with any of the following is valuable but not required:GPU infrastructure