Lead releases and production incidents. Diagnose failed deploys, capacity problems, and dependency failures with the service owner, and close them with a change in infrastructure, charts, playbooks, or the pipeline.
Bring new services, including AI platform components, onto the same release, security, and observability bar as the core products, on whichever cloud they run.
Own observability and cost: metrics, logs, and traces that show whether a release is healthy, and a clear view of spend by environment and cloud.
Hold the security baseline on the platform: identity and access, policy and runtime controls, image and dependency hygiene, and patch cadence for clusters, nodes, and hosts.
Advance disaster recovery and environment consistency so failover and drift are operating practice, not a side project.
- Set patterns other engineers can follow, and review risky changes to shared infrastructure before they land.
- Requirements
- 8+ years of experience in DevOps, site reliability, or cloud platform engineering, with several of those years leading production infrastructure.
You have led Kubernetes platforms on AWS across more than one account, including production changes you owned through the incident. You have also operated workloads on Google Cloud, Azure, or both.
You have standardized Terraform, Helm, and Ansible that other people can apply, including environment layout and secrets handling.
You have run a real deploy path for many services and understand how image build, configuration, secrets, and load-balancer registration fit together.
You can diagnose an unhealthy release: scheduling, health checks, worker backlog, and a deploy that does not become ready.
You treat IAM, network boundaries, and secrets as part of the design, and you have operated cloud key management or an equivalent.
You have run databases, cache, and shared storage under Kubernetes, including backup and a tested restore.
- You lead by setting the pattern and reviewing the risky changes, and you can pair with an engineer shipping infrastructure for the first time.
- Skills
- Cloud: AWS (EKS, ECR, VPC, ALB/NLB, RDS/Aurora, ElastiCache, S3, EFS, SQS/SNS, IAM/IRSA, KMS, CloudWatch). Working knowledge of Google Cloud and Azure for the products that run there.
Kubernetes: EKS, Helm, workloads, autoscaling, pod disruption budgets, resource sizing, and production troubleshooting.
Delivery: Terraform, Ansible, Docker, Jenkins, GitLab CI, and Git. Helm releases and Ansible playbooks across many environments.
Secrets: SOPS or equivalent, with cloud key management. No plaintext credentials in configuration.
Data: PostgreSQL, Redis or Valkey, and object and shared file storage in production, including backup and restore.
Observability: metrics, logs, and traces (Grafana, OpenTelemetry, Fluent Bit, CloudWatch, or equivalent) and cloud cost visibility.
Security: IAM, network controls, policy-as-code, image and dependency scanning, and node and host patching.
Networking: load balancers, ingress, and API gateway patterns in front of Kubernetes services.