Platform Engineer (Senior, Staff, or Principal level)

📋 Description

• • Own the developer experience end-to-end. You will design, build, and maintain SDKs, CLIs, and self-service portals that let product engineers spin up secure, cost-efficient cloud workflows in minutes—no Kubernetes PhD required.
• • Re-architect our workflow orchestration layer from the ground up. You’ll define core primitives (tasks, queues, secrets, artifacts) and wrap them in idiomatic Python and TypeScript libraries so that running a 10 000-node GPU training job feels like calling a function.
• • Harden and accelerate CI/CD. Replace brittle, 30-minute pipelines with per-merge preview environments, deterministic builds, and sub-5-minute feedback loops. Your dashboards will surface build-time, test-flake, and cost trends to every engineer and to leadership.
• • Make observability a product. Build opinionated, drop-in telemetry packages that auto-instrument services and push metrics, traces, and logs to Prometheus, Grafana, Loki, and Tempo. Create cost and reliability scorecards that turn raw data into actionable insights.
• • Codify security and compliance. Embed least-privilege IAM, envelope encryption, VPC-level isolation, and auditable secret rotation into templates and SDKs so teams inherit SOC-2-grade posture by default.
• • Champion SRE culture. Write runbooks, define SLIs/SLOs, automate incident response, and lead blameless post-mortems. Your goal: fewer 3 a.m. pages and a culture that treats reliability as everyone’s job.
• • Optimize for scale and spend. Profile multi-AZ Kubernetes clusters running 100k+ batch jobs per day, tune cluster-autoscaler, implement spot-instance orchestration, and build charge-back models that keep cloud bills predictable.
• • Mentor and multiply. As an early platform hire, you’ll set code-review standards, pair with teammates, and establish the rituals (design docs, RFCs, demo days) that let a 10-person engineering org feel like a 100-person platform team.
• • Influence product direction. Sit shoulder-to-shoulder with ML researchers and EDA engineers to understand their pain points, then translate them into roadmap items that shave hours off every experiment cycle.
• • Stay curious. Whether it’s evaluating Temporal vs. Argo Workflows, experimenting with eBPF for network policy, or contributing to open-source Helm charts, you’ll keep Quilter on the cutting edge of cloud-native tech.