Vibe-coding the MIT Course Catalog
I recently left Microsoft to join MIT's Media Lab. The transition brought an immediate problem: how do you navigate course selection when faced with the "unknown unknowns" challenge? You can easily find courses you already know you want, i.e. "known unknowns". But discovering courses you never knew existed, courses that might reshape your thinking entirely, requires different tools altogether.
MIT's official course catalog runs on what appears to be a CGI script that dates back to the 1990s. You cannot search as you type. The popular student-built alternative, Hydrant, offers decent search but displays one course at a time. Neither tool works well for browsing or screening multiple options simultaneously. More importantly, both tools remain stubbornly human-centric in an age where LLMs should help us make better decisions.
I built Courseek to solve this problem while testing how far I could push AI-assisted development. GitHub Copilot had become an expensive alternative to autocompleting functions and generating boilerplate. Could I flip the relationship entirely and dispatch goals directly to a coding LLM?
