How aryanpatankar27/prepwise-webx-ca_updated works
Project type and purpose
• This is a full-stack web application (“PrepWise 2.0”) for intelligent exam preparation:
• Frontend: React + TypeScript + Tailwind, built with Vite
• Backend: Node.js/Express + Python/Flask microservice
• Database: MongoDB
• External AI service: Google Gemini-2.0-Flash API
• PDF text extraction: PDF.js
File structure
• Top-level directories:
• frontend/ – React UI, components, pages, services (API wrappers), context, assets, templates
• backend/ – Node.js server, controllers (authController.js), routes (authRoutes.js, mcqRoutes.js), models/User.js, middleware/auth.js, config/database.js
• backend/python_api/ – Flask microservice for MCQ generation (app.py, requirements.txt)
• Patterns & layers:
• Frontend uses component-based, context for auth state, services for API calls
• Backend Node.js follows MVC (models, controllers, routes)
• A separate Python microservice handles AI and PDF processing
• Config & deployment files:
• frontend/vite.config.js, tailwind.config.js, tsconfig files
• backend/package.json, .gitignore
• python_api/requirements.txt
README for additional insights
• Frontend developer commands, environment variables (VITE_GEMINI_API_KEY)
• Backend Flask setup, MongoDB requirements, ports (frontend:5173, backend:5000, Mongo:27017)
• Features: topic-based MCQs, PDF-based MCQs, score tracking
Detail
• Break out sub-components where helpful (e.g., React Context, service modules, controllers)
• Middleware layers (auth.js) between routes and controllers
• Stateful elements (MongoDB) vs. stateless services (Express, Flask)
• Essential tech stacks and ports