As others have said, you're making a fair amount of assumptions about the way the "median" interview in Silicon Valley goes. Interviewers tend to be less concer

As others have said, you're making a fair amount of assumptions about the way the "median" interview in Silicon Valley goes. Interviewers tend to be less concerned about whether your code matches the picture in their head as much as whether your code works and isn't doing anything unduly stupid. (Being aware of performance concerns and how to optimize for best O(n) values is a huge, huge deal.) The biggest challenges I've personally noted are these: (1) Despite all the (probably sincere) claims many startups make about not looking for computer science degrees, coding questions are absolutely, positively optimized for computer science majors. I have lost more than one job because I'm "not strong enough with algorithms." For certain kinds of positions that's assuredly correct -- the engineers where I work now, who are building a high-performance realtime database, for instance -- but the number of times I've needed a routine to return a list of strongly connected components in a network graph as part of my 15-year web development career is exactly zero. (I have, however, needed it for a coding interview; as it turned out, writing that and a shortest path method and a simple parser just got me to the final face-to-face interview, where I was asked three or four more algorithmic questions. I did not get that job.) (2) I increasingly see jobs requiring experience that I literally cannot get without already having one of those jobs, i.e., "must have already worked with high-performance clusters and web sites serving millions of hits a day." It seems like the only way to get that experience is to either start a startup that gets to that point, or get an internship at a company that already has it (which is not an option once you're out of college).