Skip to content
Ahmad Asmar

About

Who I am

Section titled “Who I am”

I’m a Staff DevOps / Platform Lead with 6+ years scaling cloud-native infrastructure across AWS, GCP, and Azure. The last three of those as senior platform owner for a SaaS-scale AWS platform — built from a hybrid Azure+AWS estate into a 20-account, 4-region, 20-EKS-cluster system serving 25+ microservices to 16+ named government customers across the US and UK.

I was the sole DevOps engineer at the company for the first 18 months (May 2023 → Nov 2024), then primary IC + de-facto platform owner for the next 18. When the eventual Team Lead joined, I was onboarding him onto the stack I’d built.

Career

Section titled “Career”
WhenRoleWhere
May 2023 – present (3 yr)Senior DevOps EngineerZencity (via ASAL Technologies) — full platform ownership; AWS, EKS, GitOps, FinOps, security
Jun 2021 – May 2023 (2 yr)DevOps Engineer L2Freightos — GCP, Kubernetes, GitOps with ArgoCD, CI/CD pipelines, monitoring, DNS / TLS
Jul 2020 – Jul 2021 (1 yr)DevOps EngineerPDF Solutions — cloud infrastructure as code, configuration management, customer-facing builds
May 2019 – Jul 2020IT Support TechnicianPartners for Sustainable Development
Dec 2017 – Dec 2018IT Support TechnicianPalestine Telecommunications (Paltel) Group — first professional role

Five roles, from IT support → DevOps → senior platform owner. The pattern across all of them: own the system end-to-end, automate the toil, document for the next person.

What I work on

Section titled “What I work on”

The shape of my work the last 3 years:

  • Multi-account AWS at scale — 20 SSO-managed accounts, Transit Gateway-only routing, IAM Identity Center via JumpCloud, SCPs for guardrails
  • EKS as a platform — 20 clusters, Bottlerocket, Karpenter, Pod Identity, fleet-wide upgrades through 6 k8s versions with zero rollback commits
  • GitOps engine — Terraform → ArgoCD migration, ApplicationSet multi-cluster fan-out, dedicated devops clusters, cross-account cluster onboarding without bearer tokens
  • CI/CD platform — services-release unified template, self-hosted Spot runner fleet (open-sourced as a Terraform module), Modern Terraform CI/CD with 37% pipeline speedup
  • Service mesh + Gateway API migration — designed 12-phase Istio Ambient rollout, shipped 5 phases to all staging clusters in a 2-week sprint, validated cross-account TargetGroupBinding pattern that removes NLB hop
  • FinOps — data-driven sizing decisions, Graviton migrations, Spot fleets, gp3 / gp2 transitions, vendor renegotiation (Datadog SKU restructure)
  • Compliance / security — SOC 2 2025 (full engineering auditee scope), ISO 27001 lead responder, TanStack supply-chain attack audit (11 repos cleared in hours), WAF tuning, fleet-wide Kyverno policy
  • Open source — Terraform Registry module (3,767 downloads), 158★ Claude Code skills marketplace

Style

Section titled “Style”
  • Primary-source bias. I trust git log, AWS API responses, and kubectl get over what a doc says it should be. The hardest production problems are always the ones where the doc and reality disagree.
  • Validated rollouts. New patterns get piloted on a single cluster first. Then one staging cluster. Then the rest. Then prod. Each step proves the next. (Cross-account TGB shipped to six staging clusters in one day after a 1-day pilot.)
  • Sequencing matters. Before-compute on EKS addons; cert renewal before webhook reload; pod readiness gate before rolling deploy. Most production bugs are race conditions in disguise.
  • Built for the next engineer. Every reusable pattern I’ve shipped has either become a Claude Code skill or a Terraform module so the next engineer doesn’t have to re-derive it.

Certifications

Section titled “Certifications”
  • CKA — Certified Kubernetes Administrator (CNCF)

Now

Section titled “Now”

Open to Staff / Senior DevOps · Platform Engineering · SRE roles.

  • Remote-friendly · based in GMT+3
  • Comfortable across AWS (primary), GCP integration touchpoints, and multi-cloud Kubernetes
  • Particularly interested in platform-engineering / IDP work, GitOps platforms, multi-account / multi-tenant infra, and anywhere “build a paved road for the rest of engineering” is the brief

Reach out

Section titled “Reach out”