I think the real reason for using microservices is so that people get to play with K8s and put that on their resume. I get the "multiple teams" argument, but I

I think the real reason for using microservices is so that people get to play with K8s and put that on their resume. I get the "multiple teams" argument, but I don't think it's an architectural feature. To get multiple teams working on separate-but-connected code, you need to define the interfaces between those separate chunks of code. That's enforced by microservices, sure, but there's nothing stopping you from defining an immutable interface in the monolith too.