Kubernetes and SRE engineer

About the role

Keep the marketplace cluster and the provider agents up, and cover one block of the day.

What you'll do

• Look after our Kubernetes cluster and the provider agents. Each provider runs an agent on Kubernetes, in their own network, and that agent provisions the VPS or the game server after a buyer pays.
• During your hours, take operations: a rollout to roll back, an agent that has stopped accepting orders, a node whose name no longer matches what we stored.
• Write the handoff with what is still broken, what you already tried, and what the next person should look at first.

What we're looking for

• Kubernetes in production. kubectl is a daily tool, you have rolled a deployment back, and you can read a manifest well enough to see why a pod is not ready.
• Logs and metrics. You can tell an application error from a node that disappeared, and you have written that down for the next person on the incident.
• A short account of an incident you handled on a cluster other people were using, and what you changed so it would happen less often.
• Tell us which hours you can cover. Follow-the-sun means you own one block of the day, matched to where you live.