Martin-Lof's type theory is the foundation (or inspiration) for quite a few modern proof assistants. It is also an elegant non-set based foundation for constructive mathematics.

If you are interested in programming language theory (think language design) then, yes. In some sense, this kind of stuff is impossible to avoid.

If you're an engineer focused on shipping product, probably not. It's not TERRIBLY useful for most day to day coding tasks.

I would argue some understanding of theory is absolutely necessary if you want to make any significant tide change in CS. It's just that most people won't (myself included).