We Built an AI Medical Analyst in a Weekend at the Caltech Longevity Hackathon

We built a web app that turns unstructured medical documents (PDFs/images) into actionable insights. Users upload a file, we extract text (pdf-parse/tesseract.js), analyze it with a structured LLaMA prompt, store files and metadata in Supabase (Storage + Postgres), and present summaries, categories, abnormalities, and references in a clean UI. There’s also a Supabase Edge Function for server-side/batch processing.