Principal Platform Engineer

Principal Platform Engineer

Career Renew is recruiting for one of its clients a Principal Platform Engineer - this is an onsite position in San Francisco. Salary range: 275-330K USD base plus benefits plus equity.

We are looking for a Principal Platform Engineer with 10+ years of experience to architect, scale, and own the core Kubernetes infrastructure powering our autonomous AI agents for scientific discovery. This is a hands-on leadership role (roughly 70% IC work and 30% management) where you'll lead a small, mighty team while rolling up your sleeves and building from day one. You must be a true Kubernetes expert who has architected and operated clusters for long-running, stateful workloads, not just short-lived inference or web servers. We need someone with high EQ who can work cross-functionally with senior stakeholders whilst managing juniors, and who has recent startup experience owning an entire infrastructure platform end-to-end.

What You Will Do:

• Architect, build, and operate Kubernetes clusters orchestrating thousands of persistent, stateful AI agents that run 12+ hours while maintaining state throughout their lifecycle.
• Design and develop custom resource definitions (CRDs) and Kubernetes operators (Kubebuilder, Operator SDK, controller-runtime) to manage domain-specific workloads like AI agent lifecycles and research pipelines.
• Own the full infrastructure platform on AWS including infrastructure-as-code, observability, networking, storage, and scaling strategies with full architectural ownership.
• Lead and mentor a small platform team (1-2 engineers initially), collaborating closely with backend, ML, and research teams to translate workload requirements into reliable infrastructure patterns.
• Establish infrastructure best practices around observability, monitoring, alerting, and incident response to ensure a production-grade environment that lets science move faster.

Tech stack: Kubernetes, AWS (EKS), Terraform, Pulumi, Custom Kubernetes Operators, CRDs, Kubebuilder, Operator SDK, controller-runtime, Prometheus, Grafana, Datadog, Infrastructure-as-Code, GitOps, Docker, StatefulSets, CSI Drivers, Service Mesh, Python, Go

Requirements: 10+ years of experience in infrastructure/platform engineering, with deep hands-on Kubernetes expertise. (Mandatory) Architected and owned Kubernetes platforms end-to-end for long-running stateful workloads (not just short-lived inference or web servers) (Mandatory) Recent startup or small-team experience with full infrastructure ownership. (Mandatory) Led or mentored a small team (1 -5); formal management experience not required. (Nice-to-have) Experience with data-intensive platforms, scientific computing, or ML/AI infrastructure (Nice-to-have) Top CS/engineering school (e.g. Waterloo, CMU, Georgia Tech) is a plus but not required if work experience is strong. Cloud infrastructure experience (AWS EKS, GCP GKE, or Azure AKS); familiarity with IaC tools (Terraform, Pulumi, or similar) and observability stacks (Prometheus, Grafana, Datadog, or similar). (Mandatory) High EQ - will work cross-functionally with senior stakeholders from day one while managing a small team. (Mandatory) Biotech, pharma, or scientific computing background. (Nice-to-have)

Why You Should Join Us

Join a $70M Seed-funded team building the future of scientific discovery. Work directly with science organizations and shape how AI transforms research workflows at unprecedented speed.

We use AI agents to uncover real science breakthroughs - our flagship agent Kosmos runs hundreds of research tasks in parallel and compresses months of work into single runs.

Join a rocket ship at the perfect inflection point. They just signed a major deal and are scaling rapidly - massive impact from day one.

Principal Platform Engineer role in SF (Dogpatch). $275K–$330K base + equity. 70/30 IC-to-management split leading a small team of 1–2 to start.