[notyourloops]

> But are you any good at analyzing a combination of problems never encountered before ?

Haven't won a Nobel Prize yet, otherwise most work is done by first examining what I know vs. what I don't, and then using the appropriate tools (search, reference -> think -> implement/test -> discuss, as necessary) to accomplish the next steps.

Regardless of what I think and what I know, or even of my abilities to solve novel problems, the industry has decided upon its entrance exams.

Many times the problems vary on the surface, but the core concepts do not, and so the iterations and combinations thereof are solvable by polling previously encountered experience (study & practice).