Title: DevOps Engineer

Location: Erbil

Available Until: 5/27/2026

Key Responsibilities:

Foundation (first priority): the items below are the immediate build.

• Build CI/CD pipelines from scratch for all Java services (build → test → Docker image → Harbor → staging → production with approval gate).
• Define the Docker build standard and base image for all Java services.
• Set up Harbor (self-hosted container registry) for all Docker images.
• Deploy HashiCorp Vault for secrets management (replacing hardcoded credentials and environment variables).
• Configure Prometheus + Grafana for application metrics and alerting across all services.
• Set up and maintain staging and production server environments on FastPay's on-premises infrastructure.
• Integrate all Java services with the existing Graylog centralized logging system.

Ongoing: the items below run continuously.

• Add a CI/CD pipeline for every new service.
• Write and test the rollback runbook for every production cutover.
• Maintain environment parity — staging mirrors production exactly.
• Monitor MinIO storage capacity, MySQL replication health, and Redis availability.
• Manage on-premises server capacity and plan for growth.
• On-call rotation for infrastructure incidents.

Requirements - Technical (must-have):

• 4–6 years of DevOps or infrastructure engineering experience.
• CI/CD pipeline authoring from scratch — not modifying existing templates: GitHub Actions or GitLab CI YAML, multi-stage pipelines, secrets handling, deployment gates.
• Docker: multi-stage Dockerfiles, layer caching, image security practices.
• Linux server administration: user management, SSH hardening, systemd, disk monitoring, process management — all FastPay infrastructure runs on Linux servers you manage directly.
• Harbor or equivalent self-hosted container registry.
• HashiCorp Vault: installation, Raft HA cluster, policy authoring, AppRole authentication.
• Prometheus + Grafana: scrape configuration, alert rules, dashboard creation.
• Nginx or HAProxy: reverse proxy, SSL termination, upstream load balancing.
• Blue/green or canary deployment patterns.
• Database operations: MySQL backup, restore, replication — no managed service safety net.

General (must-have):

• Has operated infrastructure for a system where downtime has financial or regulatory consequences.
• Has executed a production rollback under pressure — not theoretical knowledge, actual experience.