Kubectl installed and connected to your cluster? Then open the CAST AI console, select the relevant Kubernetes service, and copy the script you get there.

This step involves deploying the read-only CAST AI agent in the cluster to create its map, showing pod and node sizes and where they are located.

Once your cluster shows up in the CAST AI console, you can generate the report with a single click of a button. This step will take a minute or so.

Check your current cluster compute cost and
compare it with the recommendations from CAST AI. See how different spot instance automation and cluster scheduling options impact your cloud expenses.

If you want to share your report, you can export it in Excel and PDF formats.

When you’re ready to optimize your cluster, press the button at the top of the screen. Paste the provided script into your cloud shell or terminal to create a service account. This step will allow CAST AI to manage the selected cluster to add and remove instances and handle all associated resources.