My experience with interviewing is that what I think of as level two or three problems are presented, where you have to solve some recursive problem and give the O() notation for the implementation. This happens in each of the multiple interviews over the three to six hour interview session. If you're not extremely good at this you are going to be false negatived tf out of there right quick.
Top level problems are where the solution involves days, weeks or months of study and contemplation where the solution possibly involves a large amount of refactoring of existing code. Architectural level stuff. There is no way to test for this in an hour long interview.
When I mention refactoring, I often get a blank stare from the interviewer. I had a coworker recently who insisted on calling it "refuctoring", having no concept of it or any desire to investigate.
Don't worry, in 500 years, after everything that is now studied is well understood, there will be one or more underlying principles discovered which will simplify it all.
"Artist. It doesn't matter if they're inventing the iPad or a new irrigation system. Their work is not their work. Their work is their art. It is what they were meant to do and - in the end - it is art. Both in the creation process and in the final product. Incompatibles embrace the artist's way and follow their muse. They don't care much about market research or customer insights. They know better than both."
I see most people my age (50) "closing off", they don't want to learn anything new. I think a lot of it is learned behavior, not wanting to be embarrassed or made fun of by their peers.
Recent project - WebGL 2d library at http://gschorno.github.io/glif/ (for WebGL enabled browser).
I'm flexible and can handle a lot of uncertainty on your end, as well as help you with direction if needed.
contact - greg.schorno at gee mail competitive rate