How I Built a Self-Hosted Family AI Health Steward (Your Health Data, on Your Shelf)
TL;DR — I built and open-sourced AI Health Steward , a self-hosted, private AI health manager for families. It reads photos of lab reports with multimodal LLMs, builds a structured per-person health profile, shows trends on a dashboard, and answers health questions grounded in your actual data — all running on your own server. Privacy isn't a feature; it's the whole point. Star it on GitHub . The…
I constructed a self-hosted AI Health Steward, a private AI health manager tailored for families. By utilizing multimodal Large Language Models (LLMs), the system transforms photos of lab reports into structured data, building a comprehensive per-person health profile. This profile includes personal details, metrics, diagnoses, medications, allergies, lifestyle, family history, and data provenance.
The platform offers trend visualization with anomaly markers and clinical critical-value alerts, AI consultation grounded in real data, personalized checkup plans, and regular health summaries. It operates on your own server, ensuring privacy and preventing data monetization or loss. The architecture employs FastAPI backend, React frontend, PostgreSQL, and pgvector for efficient search.
FastAPI provides clean async handlers, while React ensures a responsive dashboard. Postgres with pgvector eliminates the need for a separate vector database. The project emphasizes cost-effectiveness, running on minimal resources, and maintaining privacy by keeping data local. The UI features a web dashboard and a Feishu/Lark chatbot for quick, on-the-go access. This open-source project is available on GitHub and encourages contributions.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
