Overview
MindTrack is an AI-powered mental health and wellness ecosystem built in alignment with UN Sustainable Development Goal 3 (Good Health & Well-being). It bridges personal longitudinal mood tracking, biometric stress analysis, and clinical therapist booking into a unified mobile and web platform.
The system features a reactive Flutter mobile client for rapid daily emotional logging and notifications, an analytical Next.js web dashboard with Recharts trends, and a secure Spring Boot service layer protected by Bucket4j rate-limiting and JWT security.
Context
Mental health tracking tools often suffer from low user retention due to tedious manual data entry, lack of actionable emotional insights, and a disconnection from professional mental healthcare providers. Furthermore, most platforms lack multi-device continuity between quick on-the-go mobile logging and comprehensive desktop analytical reviews.
Engineering
A dual-client architecture where a cross-platform Flutter application and a Next.js 14 web portal interface with a shared Spring Boot 3 enterprise REST API. Real-time trend calculation and emotional pattern scoring are persisted in PostgreSQL, with Bucket4j safeguarding API quotas.
graph TD
Mobile["Flutter Mobile App\n(Dart + Riverpod)"]
Web["Web Analytics Portal\n(Next.js 14 + Recharts)"]
Gateway["API Gateway / Security\n(Bucket4j Rate Limiter)"]
Backend["Spring Boot 3\n(Java 21 + JPA)"]
DB[("PostgreSQL Database")]
Notif["Local Notifications\n& Wellness Triggers"]
Mobile -->|"REST / JWT"| Gateway
Web -->|"REST / JWT"| Gateway
Mobile --> Notif
Gateway --> Backend
Backend --> DBStack Breakdown
| Layer | Technology | Purpose |
|---|---|---|
| Mobile Client | Flutter 3 + Dart | Riverpod state management, GoRouter, fl_chart, Dio, Lottie |
| Web Portal | Next.js 14 + TypeScript | Tailwind CSS, Recharts analytics, NextAuth, responsive web portal |
| Backend Service | Spring Boot 3 + Java 21 | Spring Data JPA, Spring Security, Bucket4j rate limiting |
| Database | PostgreSQL | Relational persistence for longitudinal logs, habits, and therapist records |
| Notification Layer | flutter_local_notifications | Scheduled mindfulness habits and emotional check-in nudges |
Features
Engineering depth
The Problem
Users frequently switch between recording mood on their mobile device and reviewing weekly trends on their desktop browser, requiring strict timestamp synchronization.
The Solution
Established standardized UTC millisecond timestamp schemas and synchronized JWT session refresh tokens across both client platforms.
The Problem
Calculating dynamic rolling averages, stress variance, and streak metrics across months of entries caused slow database queries.
The Solution
Designed compound indexing on user_id and recorded_at with pre-aggregated weekly summary views in PostgreSQL.
Reflection
Single-tap mood sliders and frictionless mobile interactions drive significantly higher daily habit completion rates.
Rate limiting with Bucket4j at the gateway layer is critical for protecting mental health logging endpoints against abuse.
Aligning projects with measurable social frameworks like UN SDG 3 sharpens product scope and user value.
Continue Reading
Web App
Unified Agile sprint planning, Kanban workflow, and delivery tracking workspace.
AI/ML
AI-powered grant & scholarship intelligence platform with semantic matching and automated application drafting.
Web App
Unified hospital management ecosystem for clinical and administrative workflows.