← Back to work

Atlas Copco Group
Platform & DevOps Engineer
Oct 2025 – Present  ·  Antwerp, Belgium
Atlas Copco
Industry Software Cloud
cpu %78.3%liverequest latency (p99)SLA124msavgmemory allocation6.2 GB/ 8 GBdeploy-prodbuild-mainlint-checkstaging-synctest-suitepod restartdisk 85%

terraform plan · 0 to add, 2 to change, 0 to destroylast sync: argocd/prod-cluster

Role

Cloud observability, infrastructure, and platform tooling for a distributed Industrial IoT environment serving six cross-functional sub-teams — spanning both the global and Chinese Azure tenants, which run as separate clouds and have to be instrumented separately.

What I did

  • Built observability across six IoT sub-teams — New Relic dashboards, NRQL alerting pipelines, deployment markers, and ingestion alerts for error triage, third-party API failures, and per-resource metrics — all provisioned through Terraform.
  • Wrote Azure Function Apps in Python to collect and push custom metrics from both the global and Chinese tenants into New Relic, extending a single observability picture across two clouds that do not talk to each other.
  • Migrated UltraDNS management to Terraform, removing manual steps from the release process, and consolidated cross-team cost reporting across 60+ resource groups so business units could see their own spend.
  • Took a Kubernetes migration from proof of concept into development: set up the clusters and a GitOps delivery pipeline on ArgoCD, Helm, Kargo, and Azure DevOps, with multi-environment promotion and approval gates.
  • Deployed databases and services onto those clusters — including ClickHouse across the global and China environments — with the supporting ACR mirrors, registry proxies, private networking, and subnets.

Tech stack

Python Terraform Kubernetes ArgoCD Helm Kargo GitOps Azure Function Apps Azure Pipelines Azure Container Registry ClickHouse New Relic NRQL UltraDNS SQL