I save £300/mo by auto-scaling my staging cluster at night

We have a staging environment running 24/7 — 6-7 microservices sitting idle every evening and weekend. That's roughly £300/month wasted.

I wanted a simple automation: scale deployments to zero at 8pm, restore at 7am. The options were cronjob+kubectl scripts. (fragile, no visibility) or writing a custom controller (kinda overkill for this).

This drove me to build something which became Tiny Systems — an open-source visual workflow engine where automations run as native K8s operators.

The cost saver flow uses cron triggers, a deployment lister filtered by labels, a scaler, and Slack notifications.

Here's the ready-to-install solution: https://tinysystems.io/solutions/cluster-cost-saver-0

Solo project, happy to answer questions.