MealCircle — Clinical Nutrition SaaS

MealCircle keeps plans, food logs, check-ins, and messages in one shared care record so patient drift shows up early — while there's still time to follow up. A 22-domain Python backend, a Next.js expert web app, and an Expo patient app on a generated SDK, built and operated solo.

Live

Mobile App

9,768

Food Catalog

22 Domains

Backend

Case Study Analysis

The Challenge

Nutrition care falls apart between appointments: providers usually find out a patient has drifted at the next visit — or never.

The Solution

One care record shared by the expert workspace and the patient app, with retention signals that surface who needs attention today, a 9,768-item food catalog, and an AI import that turns an existing meal-plan document into a reviewable draft.

Key Features

Retention Signals: Rule-based health states surface at-risk patients early.

Patient 360: Plans, logs, biometrics, and messages in one record.

Modular Monolith: 22 backend domains, ~47k lines, 63 test files.

AI Plan Import: Claude-based structured extraction from meal-plan documents, human-reviewed.

PHI Security: AES-256-GCM encrypted patient data with an immutable audit trail.