A recipe recommender that structures raw recipe data with NLP and clusters dishes by theme and flavor profile, so users can ask in natural language and get personalized suggestions. Took third place as a practical showcase of text-based workflows.
Context matters here: this was a hackathon about the free tier, not a contest of large-scale architecture. What that format rewards is showing what can be done inside modest limits, and a clean pipeline from messy raw data to a working recommendation fits the brief. Choosing recipes helps too, because anyone can judge on the spot whether a suggestion is sensible — the output verifies itself in front of the panel without extra explanation.
Summaries are written by this site. Project pages include demo videos, screenshots and the team's own write-up (videos may autoplay).