I would love to see how Qovery or other orgs organize projects, environments, and clusters internally. I’ve always wondered what this might look like at companies of different scales running different types of workloads (internal tools vs user facing applications vs data eng tools, etc.)