Dstny

From scratch, on two runtimes at once

Terraform-based, on AWS EKS, with GitOps workflows and GitLab CI, then adapted to run concurrently on OpenShift. Supporting two Kubernetes distributions in parallel constrains every choice you make about ingress, storage and security policy, and it rules out most of the shortcuts a single-runtime platform can take.

Runtimes

EKS + OpenShift

On call

Throughout

Fifty customer domains, no manual certificates

AWS ALB and NLB front a Traefik ingress; cert-manager issues and renews certificates for 50+ end-customer custom domains, with an in-cluster DNS service so onboarding a new one takes minutes rather than a ticket. Prometheus, Loki and Grafana run as a self-service stack the application teams reach into themselves.

The white-label login is a product, and non-engineers operate it

Authentication runs on a redundant Keycloak deployment. Beside it I built a standalone white-label theming service in TypeScript and React, used directly by the onboarding and account management teams across 10+ enterprise deployments, with no engineer in the loop to rebrand a customer's login. I also wrote the training material and ran the Kubernetes sessions with the operations teams who inherited all of this.