The biggest flaw is the learning loop. When you encounter a new architectural or algorithmic concept, you are forced to break your flow. You have to leave the platform, read abstract documentation, watch a 3-hour YouTube video, or sift through AI slop. By the time you return to the editor, you've experienced massive context switching.
This applies to tech education as a whole: we find information, read it, take notes, and then never look at those notes again because passive re-reading is boring. As a result, we forget most of it.
I wanted to fix this feedback loop for myself and others. I built Self University — a platform that turns everything you can be asked in an engineering interview into an interactive puzzle that you assemble by hand. It features a massive knowledge graph covering hundreds of puzzles across networking, databases, cloud native, backend cores, frontend performance, and architecture.
It’s built on the Generation Effect. Instead of reading dry text or fighting raw syntax overhead, you visually build the structure in your head by solving concrete interactive puzzles. To make it more engaging, I also added a PvP Arena where you can challenge others to speed-running these architecture puzzles. It's fast, interactive, and completely live. https://selfuniversity.site