Devops Engineer Mid Level

## About the Role We are hiring a mid-level DevOps Engineer to help own the infrastructure and delivery surface of a leading technology marketplace that has been running successfully for the last 10 years. The infrastructure is a multi-region cloud estate running dozens of microservices on Kubernetes, an edge router, a Node.js proxy fleet, a Kafka event backbone, a data plane built on ClickHouse, Spark, Snowflake, and Airflow, and modern CI/CD pipelines. The job is to keep that estate reliable, cheap, secure, and pleasant to ship into. As the platform adds agent-native traffic, you will help ensure the infrastructure scales for machine-speed, high-frequency agent calls alongside human traffic — without separate handling or runaway cost. ## Roles and Responsibilities - AWS infrastructure — Operate and improve the multi-region AWS estate across compute, streaming, databases, caching, and storage. Diagnose issues across regions and accounts. - Infrastructure-as-Code — Maintain and extend the Terraform modules that define the platform. Treat infra changes like code: reviewed, tested, deployed via pipeline. - Kubernetes — Operate workloads on Kubernetes: deployments, autoscaling, ingress, networking, secrets, resource limits. Help service teams adopt safe defaults. - CI/CD — Own and improve the CI/CD pipelines that ship frontend, backend, and data-platform code. Keep build times short, test suites fast and reliable in CI, and deploys boring. - Observability platform — Own the observability and alerting platform and the standards service teams instrument against. Service teams own their own instrumentation — you own the platform that makes it possible, the conventions that keep it consistent, and the gap analysis that catches what nobody instrumented. Take part in the on-call rotation. - Incident response — Lead or co-lead incidents; run blameless post-mortems; turn each incident into automation or improved runbooks. - Security and access — Manage secrets, rotation, access policies, and network policies. Partner with the architect on zero-trust patterns, OAuth2 / JWT / mTLS, and security compliance. - Cost discipline — Hunt down idle capacity, oversized clusters, runaway warehouse queries, expensive streaming shards, and CDN waste. Report monthly on what came down and why. - Vendor consolidation — Help simplify the observability, CDN, CDP, and supporting vendor stack. Carry through the migrations. ## What Success Looks Like - Uptime ≥ 99.99% across the platform; incidents are rarer, shorter, and better understood than before you joined. - Deploys to every part of the stack are routine — service teams ship multiple times a day without your help. - Terraform is the source of truth: nothing important is configured by hand in the console. - Alert volume is down and alert quality is up — on-call does not dread the pager. - AWS spend and vendor spend trend down per call as traffic grows. - Security and compliance audits go smoothly because the controls are real and the evidence is automated. ## What You Bring - 5+ years of hands-on DevOps / SRE / Platform Engineering experience in a production environment that mattered. - Strong hands-on AWS across compute, networking, storage, and managed databases. - Kubernetes in production — deployments, autoscaling, ingress, helm, secrets, RBAC, troubleshooting. - Terraform in production — modules, state hygiene, code review discipline. - Solid scripting in Python, Bash, or Go. - CI/CD experience with modern pipeline tooling. - Real observability experience — instrumentation, dashboards, alerts. - Comfortable on-call. Comfortable running an incident. Comfortable telling the team that something needs to be redesigned. - Security mindset: least-privilege, secrets hygiene, supply-chain awareness. - Bonus: experience operating Kafka, ClickHouse, Snowflake, or Airflow; exposure to agentic / LLM-driven ops workflows for incident triage or remediation. - You automate before you complain. You leave the cluster cheaper and quieter than you found it.