AI Specialist Finder

A web platform that connects users with specialised professionals nearby, combining location-based services with performance metrics so the ranking reflects evidence rather than advertising spend.

Why I built it

The idea came from a specific failure. While rock climbing in Vietnam I split my knee open and needed urgent help, but I could not communicate with anyone locally and had no way to find the right person quickly. Afterwards it seemed obvious that the hard part was not the map. It was ranking who was actually good.

How it works

The application integrates the Google Maps, Places and Geocoding APIs for location-aware search and filtering, with an OpenAI integration handling natural-language interaction. On top of the raw results sits a ranking system that scores professionals using review-derived performance indicators, gathered through a custom search engine, so a quantitative signal drives the ordering.

The frontend is React, the backend Flask, and it is deployed on Heroku with keys held in environment variables rather than in the codebase.