this was exactly my point, if you are lucky and joined company which has high potential to grow and/or joined Google like companies with crunching lots of compe

this was exactly my point, if you are lucky and joined company which has high potential to grow and/or joined Google like companies with crunching lots of competitive programming problems (leetcode, hackerrank) then you might join teams where people work on kubernetes, large scale storage, low latency proxies and so on. Even though some of them are open source, it is not easy to do contributions after whole day of coding. I did this in the past, to become productive in projects like k8s you should invest a lot of your own time. At small scale k8s rarely fails and features out of box is enough for most companies, since you don't have large scale problems/experience it is difficult to come up with feature which could be useful for k8s community. this applies to other projects as well. Unless you are working full time on field, it's not easy to become expert in that field