When we reach for it
Estates with enough services that the scheduling, rollout and self-healing behaviour is doing real work — and clients who want to avoid depending on any one cloud’s proprietary runtime.
Kubernetes engineering
The industry’s standard scheduler for containers, and the reason a workload can be described once and run the same way in every environment.
What we reach for first
When we would argue against it
It is a poor answer for three services and no platform team. We have talked more clients out of Kubernetes than into it; a container service or a managed runtime is frequently the correct, unglamorous choice.
What it looks like in delivery
Where it is justified, we treat the cluster as a product with users: sane defaults, resource limits that exist, deployments that roll back on their own, and enough documentation that the client’s team can run it without us.
Where this appears on the site
Published work that names it
Derived from the stacks published on those pages, not written here — so this list cannot claim something the page it points at does not say.
Also in cloud and infrastructure
Working in Kubernetes?
Tell us what it is running, what it costs you today, and what you need it to do next. A senior engineer will tell you what we would keep and what we would change.