StaffEngineer,AIInfrastructure&MLOps

## **About the team** The MLOps Platform team builds and operates Pfizer’s self-service MLOps platform. We give researchers, ML scientists, and application developers a paved road to discover, validate, register, deploy, and consume models under enterprise security and compliance constraints. ## **What you’ll do** * Contribute to the evolution of the platform through design documents and implementation of cross-cutting improvements across validation, serving, registry, and infrastructure. * Drive our **multi-cloud and Kubernetes expansion** (AWS today, GCP and GKE next), keeping both runtimes coherent behind shared abstractions for deployment, observability, auth, and audit. * Own the managed **serving and registry lifecycle** end-to-end, including network-isolated validation, approvals, and integration with a shared GenAI gateway. * Own the **infrastructure, build, and runtime stack** : Terraform across providers, container orchestration, CI/CD, and a vendor-agnostic observability and on-call model. * Take end-to-end ownership of features and platform components, from requirements gathering and design to implementation, deployment, and production support, contributing to overall engineering quality. ## **What we’re looking for** * Senior platform/backend experience, with multiple years owning a production system end-to-end. * Strong **Python** skills, with enough frontend literacy to review and contribute across the stack. * **Kubernetes in production** and **Terraform** across multiple environments and providers; experience with at least one of **AWS** or **GCP** (and ability to learn the other). * A track record of writing design documents, ADRs, and runbooks — not just code. ## **Nice to have** * Hands-on experience operating **ML serving or model registry systems** in production. * Experience evolving a platform from single-cloud to multi-cloud without forking the product. * GitOps (ArgoCD / Flux), hermetic monorepo builds, and shared GenAI gateway integrations