What I actually look for in a portfolio project

I used to think a portfolio project had to be large to count. Now I care more about whether it solves a real problem and whether I understand the decisions inside it.

My current bar

I want a project to have a clear purpose, instructions that work, and some way to verify the important behavior. A few focused tests are more useful than a long feature list I cannot defend.

The part that matters in an interview

I should be able to explain what I left out, what broke while I built it, and what I would change next. That is the part that shows whether the project taught me anything.