We've been using it with a ~10 nodes cluster for some time to host both internal and production apps. It has allowed the rest of the team to focus on the apps

We've been using it with a ~10 nodes cluster for some time to host both internal and production apps. It has allowed the rest of the team to focus on the apps themselves without worrying about how to deploy them, how to provision more capacity, how to rollback in case of failure, etc. Using the builtin affinity and priority system, we were able to make sure that production apps would always run smoothly without being affected by our other processes.