[clickonchris]

I'm with the poster. After 8 years of professional experience I'm great at building large, stable systems that work, but I'm not even that interested in complex algorithms.

For me it is a question of cost/benefit analysis. What real value do you get out of being able to write super mathy algorithms when the solution is often a Google search away? More importantly, programmers get paid to build software that works, not to recite algorithms. (that's a generalization. Google & Amazon probably care much more about in-depth knowledge of algorithms but I consider them the exception).

Take the search algorithm example. Don't pre-optimize. I'll use the simplest linear search I can to get the job done, and if needs be, look up a faster algorithm to speed up the search.