Job Description:
Experience: 8 to 12 years overall, at least 3 years in DevOps / platform engineering on AWS and Kubernetes
About the role
Our platform is a set of Java 21/25 / Spring Boot microservices, batch jobs and React micro-frontends on AWS EKS. Several agile teams ship to a shared estate on weekly release lines.
We want an engineer who owns the path from commit to production: build pipelines, infrastructure as code, Helm deployments, release management, and security remediation. You should also be able to read and change application code when a build or deploy breaks. Over the next 12 to 18 months you will lead our CI/CD migration from AWS CodePipeline / CodeBuild to Harness.
What you will do
CI/CD and the Harness migration
Run and improve our current pipelines: AWS CodePipeline / CodeBuild buildspecs, with pipeline definitions written in AWS CDK (TypeScript).
Design and deliver the move to Harness CI/CD. That means pipeline templates, environments, approval gates, secrets, and canary or rolling deployments for about 10 microservices, nginx ingress, CronJobs and the SQL migration pipeline.
Run the old and new pipelines side by side during the migration, with a clear plan to cut over and roll back.
Kubernetes and AWS infrastructure
Maintain the Helm charts (a shared library chart plus per-service charts) for services, CronJobs and the nginx gateway on EKS (Fargate).
Write and review Terraform (Terraform Enterprise) for Aurora PostgreSQL, API Gateway, SNS / EventBridge, IAM and networking.
Troubleshoot deployment failures, for example pods stuck Pending, Helm --atomic rollbacks, security-group-for-pods problems, image issues, and ingress routing.
Releases and environments
Run weekly release trains across many repos. That covers cutting release branches, building the release manifest, checking image tags against what is deployed, sequencing deployments when one service depends on another, and supporting change requests for production.
Keep the environments (dev, QA, staging, prod) consistent and promote changes between them in a predictable way.
Security and compliance
Triage and fix findings from Wiz, Veracode and Prisma (container and dependency CVEs). Know the difference between direct and transitive dependencies, and how to roll out a version bump across many repos.
Work with hardened, shell-less base images (Chainguard-style), CyberArk-managed secrets, and Okta-based service authentication.
Full stack support
Debug and patch Java / Spring Boot (Maven) services, Python batch jobs, and Node / React builds when pipelines or runtime issues need it.
Support database change delivery (Flyway / SQL migrations on Aurora PostgreSQL).
Add monitoring in Datadog (logs, metrics, monitors, SLOs) and help during incidents and postmortems.
Automation and AI tooling
Script repetitive work in Python, Bash, or TypeScript: deploy helpers, scaling, release checks, commit audits.
Use AI-assisted engineering tools (for example Claude Code or Copilot) responsibly to speed up pipeline and infrastructure work.
Must have
Strong hands-on AWS: EKS, IAM, VPC / security groups, ECR, Aurora / RDS, API Gateway, CloudWatch.
Production Kubernetes and Helm: writing charts, debugging failed rollouts, Fargate or managed node groups.
Terraform at a professional level: modules, state, plans in code review.
Has built enterprise CI/CD on at least one of Harness, Jenkins, GitHub Actions or AWS CodePipeline. Harness experience is a strong plus.
Docker image builds and image hardening, and a working understanding of container security scanning.
Good Git practice: branching strategies, release branches, pull-request-based promotion.
Able to read and debug Java / Spring Boot and at least one of Python or TypeScript / Node.
Clear written communication: runbooks, release notes, and incident updates.
Nice to have
Harness CI/CD certification, or a previous migration onto Harness.
AWS CDK (TypeScript).
Datadog, Wiz, Veracode or Prisma Cloud.
Flyway or Liquibase database migration pipelines.
Experience in financial services or another regulated industry: change management, separation of duties, audit evidence.
Worked in a multi-team agile setup (Jira, Confluence).