System Architecture

Overall User Flow Of The App

Here’s a shortened intro for Flinder along with clear steps to set up the project:

---

🏠 Flinder – Smart Roommate & Flat Finder

Flinder is a modern roommate matching app that uses AI, real-time chat, and a swipe-based interface to help users find compatible flatmates and apply for flats as a group. The platform blends Supabase for auth and real-time chat, Express.js for API orchestration, and a Flask-based ML engine for compatibility scoring based on user bios and lifestyle tags.

---

⚙️ Project Setup Guide

1. Frontend: Flutter App

Handles UI, user interactions, swipes, and real-time messaging.

• Install Flutter: Flutter Docs
• Clone the repo:

  git clone https://github.com/your-org/flinder-app.git
  cd flinder-app
  flutter pub get

• Add environment config (.env) with:

  SUPABASE_URL=your-supabase-url
  SUPABASE_ANON_KEY=your-anon-key
  EXPRESS_API_URL=http://your-express-server

• Run on emulator or device:

  flutter run

---

2. Backend: Express.js API Server

Orchestrates swipes, groups, chat logic, and talks to Supabase & ML engine.

• Navigate to backend:

  cd flinder-backend

• Install dependencies:

  npm install

• Create .env:

  SUPABASE_URL=your-supabase-url
  SUPABASE_SERVICE_KEY=your-service-role-key
  ML_API_URL=http://localhost:5000
  PORT=3000

• Start server:

  npm run dev

---

3. ML Engine: Flask + Docker

Extracts tags from bios, calculates compatibility scores.

• Navigate to ML service:

  cd flinder-ml

• Build & run Docker container:

  docker build -t flinder-ml .
  docker run -p 5000:5000 flinder-ml

---

4. Supabase Setup

Manages auth, users, chats, swipes, groups, and real-time messaging.

• Create a project on Supabase
• Enable:
  • Auth (email/password)
  • Realtime for messages, swipes, groups
• Run SQL to create tables:
  • users, swipes, chats, messages, chat_members, groups, flats
• Get your SUPABASE_URL and keys from project settings.