To a point, but like most things in software my experience is you look up someone who's already figured it out and copy them :) Granted you have to understand the code a little more and be able to apply it to whatever system/hardware constraints you're working with.

But just as higher level Software Engineers rarely have a need to implement recursion/graph traversal/etc, in embedded its even rarer due to the relative expense of such operations. I'd say knowledge of signal processing and how to power/performance/memory-optimize code would in general be much more valuable.