I mostly agree with you, but the side project can be a huge factor in differentiation between candidates, either at the resume or interview stage. My rules for

I mostly agree with you, but the side project can be a huge factor in differentiation between candidates, either at the resume or interview stage. My rules for the ideal portfolio ready side project: 1. It is "complete". Not just started but something that actually works. This means that a side project needs to be something which you can actually finish. No matter how great your code looks, it will be outweighed by code that works. 2. The problem it can solve can be described in 10 seconds. 3. The interviewer will not be able to sketch it out in their mind in under a minute. While the problem solved may actually be quite simple to implement, it is best if it's just outside the interviewer's knowledge. Actually publishing a usable library, even something trivial, provides great separation at the resume level and gives you something specific to discuss in the interview. The easiest examples I could think of would be a library that does some math. Odds are the interviewer doesn't know how to do trig functions on a sphere, but it is something easy to look up and implement. You can describe it quickly. No amount of time will help them figure it out. Some people's eyes will just glaze over and they'll assume you're really smart (don't go into too much detail on these guys, show them your communication skills instead).