Context
Sourcing and large trade events drown people in options. The information is technically available but practically unsearchable. Thousands of suppliers, booths, and SKUs with no good way to find the right one.
Problem
Discovery at scale is a navigation problem disguised as a search problem. Keyword search fails when you do not yet know the right words, and a map of an expo floor does not tell you which booth is worth your time.
What I did
- Prototyped an AI interface that lets people describe what they need and navigate suppliers and events conversationally.
- Focused on turning unstructured, messy real-world inventories into something queryable.
What I learned
- AI is most useful exactly where the search box gives up.
- Discovery problems are really 'help me know what to ask' problems.
This started as a practical itch. Trade events and supplier catalogs are full of value and almost impossible to navigate efficiently.
From keywords to intent
The interesting move is letting people describe intent instead of guessing keywords, then doing the work of mapping that intent onto a messy real-world inventory. It is a small project, but it sharpened my view that the next interfaces are conversational and context-aware by default.