I found the HashiStack to be manageable by a single person or very small teams. It's easy to set up and hasn't let us down so far. We get away with using shell scripts to test and build our containers for now. The trickiest part is scaling as soon as it becomes necessary. Our current three-node cluster handles a few hundred requests a second without issues, but I migrated the database three times already, and I'm in the progress of migrating it a forth time to a much bigger machine.

You can find more about our setup and ops here: https://pirsch.io/blog/techstack/