Some fields might be locked up by postgrads (crypto engineering is not one of them, and I doubt language tooling is either). You should check before you invest too much in it. But I would advise against being spooked by academic credentials. Good places to work care more about conversance than credentials, and you get to draft off of all the research work the academics publish before moving into industry to compete with you.
I've been in this industry since copper and telco was hot shit.. I went from doing Telco programming > IT > Web Development > scalable enterprise applications.
Why bother stressing over something you have no control over? Do what you love and the money follows.
i do have some hope that i've gotten into my current niche (static type inference for dynamic languages) reasonably early in its life cycle, but if i'm wrong i don't want to have not kept up with the larger world at least to some extent.
(to be precise, i'm not so much worried about being able to transition my skills to something else, as about not being given a chance to do so unless i can already demonstrate the new skillset people are looking for.)
i've been wondering if i should pick up something entirely different on the side (e.g. machine learning for broad applicability, though that does not excite me at all), or to instead diversify from my current niche but learn to write tools across a broader variety of languages and ecosystems. it's scary sometimes how the entire programming ecosystem can change overnight and leave once-valuable skills in the dust.