FlareScan by Yash Nair
FlareScan is an independent web app for people with recurring inflammatory skin concerns. It brings reported symptoms and everyday routines together to help users explore possible connections and prepare clearer questions for a clinician visit.
Personal insights, with context
The assessment covers food and drink, sleep, stress, skin and hair products, environmental exposures, and care or medicines. Users report whether a clinician has confirmed their condition and can separately consent to optional photo review. Claude models consider the answers to select one experimental lifestyle-pattern score and explain possible connections, competing explanations, and uncertainty.
The score does not measure disease severity or establish a diagnosis. FlareScan is a research prototype and has not been clinically validated. Continue prescribed care and seek professional advice for medical decisions.
The project
Yash Nair created and directed FlareScan as an independent project, using AI assistance during development and testing. The app uses TypeScript, JavaScript, React, Next.js, and Anthropic’s Claude API. Its models are specialized through instructions and reference material, rather than custom-trained model weights.