Luminary Cloud helps engineering companies be more competitive by getting to market faster, creating new, better products, and reducing development risk. We do this with our Physics AI platform, the fastest and easiest way to build and deploy models to understand and instantly predict physical reality with precision. Customers span industries from automotive and aerospace, to leading sporting equipment providers, including Otto Aviation, Joby Aviation, Piper Aircraft and Trek Bikes. Luminary is a Series B company and is headquartered in San Mateo, California.

About the role

We’re building out a cloud platform team and looking for a Senior DevOps Engineer to own the developer infrastructure that powers our products. You will own how we deploy, scale, observe, and secure systems across GCP and AWS, with Kubernetes at the core.

This isn’t a ticket-queue role. You’ll work directly with engineers building services in Go and TypeScript, researchers training PyTorch models, and leadership defining the roadmap. You’ll have real ownership and the latitude to build things the right way from the start.

What you’ll do

• Design, build, and operate cloud infrastructure on GCP with an emphasis on reliability, security, and cost efficiency
• Own and evolve our Kubernetes platform — cluster architecture, RBAC, networking, autoscaling, and workload scheduling
• Build and maintain automated CI/CD pipelines using GitHub Actions and ArgoCD, supporting GitOps workflows for all services
• Write Go and Python tooling to automate infrastructure tasks, improve developer experience, and extend internal platform capabilities
• Establish observability practices — metrics (Prometheus/Grafana), distributed tracing (OpenTelemetry), and centralized logging
• Define and enforce security best practices: secrets management (Vault/KMS), image scanning, IAM least-privilege, and network policies
• Support GPU-based ML workloads, working with researchers to provision and optimise node pools for PyTorch training and inference
• Respond to incidents and lead blameless postmortems to drive continuous improvement in system reliability
• Write clear documentation and champion a culture of engineering excellence across the team

What we’re looking for

Required

• 5–8 years of experience in DevOps, SRE, or platform engineering roles
• Production Kubernetes experience — cluster management, not just deploying workloads
• Hands-on experience with GCP or AWS; solid conceptual understanding of both
• End-to-end ownership of CI/CD pipelines and GitOps workflows
• Proficiency in Go or Python for writing infrastructure tooling and automation
• Infrastructure as Code expertise with Terraform or Pulumi
• Experience with observability stacks: Prometheus, Grafana, and a log aggregation platform
• Strong grasp of cloud security fundamentals: IAM, secrets management, network policies

Preferred

• Experience supporting ML training infrastructure, GPU node pools, or model serving (TorchServe, Triton)
• Familiarity with TypeScript for build tooling or internal developer platforms
• Background in a fast-moving startup or product engineering environment
• Contributions to open-source infrastructure tooling

Certifications

The following are valued but not required:

• Google Professional Cloud DevOps Engineer
• CKA or CKAD (Certified Kubernetes Administrator / Application Developer)
• AWS Solutions Architect or DevOps Professional
• HashiCorp Terraform Associate

Our tech stack

Why join us

• Work on a development team building frontier AI models for Physics AI
• Collaborative, low-ego team that values craft and clear thinking
• Competitive salary, equity, and benefits