The smartest people gravitate, by taste and economics, to hard problems that aren't already solved by existing abstractions. For the vast majority of software

The smartest people gravitate, by taste and economics, to hard problems that aren't already solved by existing abstractions. For the vast majority of software project, it's sufficient to rearrange existing abstractions in ways consistent with their intended use. When you hit a legitimately hard problem, it's may offer an exciting chance to gain a real technological advantage on the competition, but you have to make damn sure that the competition won't blow by you because they found a way to not solve the hard problem.