Site Reliability Engineer

About The Role

The role owns the reliability of production infrastructure serving millions of requests per day - Kubernetes clusters, distributed services, and the observability stack that keeps them honest.

You will sit at the intersection of infrastructure engineering and incident response: automating away toil, designing SLOs that actually mean something, and being the person the on-call rotation trusts when things break at 3am.

Key Responsibilities

• Design, build, and operate containerized infrastructure on Kubernetes (EKS/GKE), including cluster upgrades, autoscaling policies, and multi-environment rollout strategies
• Build and maintain infrastructure-as-code using Terraform and Helm, with CI/CD pipelines in GitHub Actions or GitLab CI for infrastructure changes
• Define and enforce SLOs/SLIs with error budgets; build dashboards and alerts in Prometheus, Grafana, and Datadog that page on real problems, not noise
• Lead incident response for production outages - triage, mitigation, root cause analysis, and blameless postmortems with actionable follow-through
• Automate away recurring operational toil: runbook automation, self-healing infrastructure, and capacity planning scripts
• Harden production systems for reliability and security - chaos engineering experiments, disaster recovery drills, and failover testing
• Partner with application engineers on service design reviews, ensuring new services ship with proper observability, load testing, and rollback paths

What We Are Looking For

• 3-7 years of experience in SRE, DevOps, or infrastructure engineering, including significant on-call ownership of production systems
• Deep hands-on expertise with Kubernetes in production - operating clusters, debugging workloads, and tuning resource limits and autoscaling behavior
• Strong infrastructure-as-code skills with Terraform (or equivalent) and experience managing cloud infrastructure on AWS, GCP, or Azure
• Production experience with observability tooling: Prometheus/Grafana, Datadog, or OpenTelemetry, including writing PromQL/queries from scratch
• Proficiency in at least one systems language: Go, Python, or Bash, with a track record of building internal tooling
• Solid grasp of distributed systems fundamentals: load balancing, caching, DNS, TLS, queuing, and failure modes at scale
• Bonus: experience with chaos engineering tools (Litmus, Gremlin), service mesh (Istio/Linkerd), cost optimization (FinOps), or running multi-region active-active architectures