Doctor AI
Full-stack AI health assistant — 23 guided analysis categories powered by Google Gemini. Users have a multi-turn conversation and receive personalised health insights. Client project.
Disclaimer: Doctor AI is an AI-powered health assistant for informational and educational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider with questions about a medical condition.
1The Problem
A client needed a production-ready AI health assistant that could guide users through personalised health analysis across multiple domains — nutrition, sleep, stress, exercise, hormones, gut health, and more — without requiring a human support team behind it.
2My Role
Sole developer. Built the Django application, the Gemini service layer with its model fallback chain, the 23 category prompts, and the frontend.
3The Engineering
Built a full-stack Django 5 application with a Google Gemini-powered AI service layer. Engineered 23 custom system prompts — one per health category — to drive targeted multi-turn conversations. Implemented dual user modes (guest in-memory sessions and authenticated persistent history), a REST API with CSRF protection, and a responsive vanilla JS/HTML/CSS frontend with Markdown rendering via Marked.js.
4The Outcome
Delivered a fully functional AI health assistant with 23 analysis domains, persistent conversation history for logged-in users, and a mobile-responsive UI. The AI model fallback chain (gemini-2.5-pro → gemini-pro → gemini-1.0-pro) ensures reliability. Production-ready with Gunicorn, environment variable config, and PostgreSQL-compatible data models.
Screenshots
Have a similar project?
I'm available for new projects — let's talk about what you need.