Platform Engineering: Compute

About the team

We are a small team of 5 people who are responsible for running Kubernetes clusters and supporting services like CoreDNS, kube-proxy, Cluster Autoscaler and others. We closely cooperate with the other Platform Engineering team called “Databases” to provide a stable platform based on AWS services with focus on self-service capabilities for other teams.

Besides that we are still improving our services, toolchains and workflows to make our working environment as comfortable as possible. Minimizing the amount of repetitive and dull work is our key for easy going work.

We keep oncalls about the production platform and so the stability of the platform is one of our highest prioritiesy. Every incident is investigated very thoroughly to avoid recurrence in the future.

How we work

Our whole platform is a very complex environment that is still very active, like nature in spring. We are constantly improving our services, applying the newest practices and keeping our platform up to date to minimize our technical debt.

A Platform Engineer in our team plays with Terraform/Terragrunt and Tanka/Jsonnet on a daily basis to interact with services deployed on AWS or Kubernetes. Our entire platform is aligned with the IaC standard so each change is properly reviewed before it gets merged into the Git repository. We try to minimize the bus factor so we regularly share knowledge across the team to improve the effectiveness of our effort.