How shaikabdulkhadar1/ai-budget-management-app works
Project type and purpose
• This is a cross-platform mobile full-stack application (“AI-Budget Tracker”) built with React Native/Expo on the frontend and a cloud backend (Firebase / Supabase) plus external AI/OCR services.
• Purpose: let users track income/expenses, set budgets, get AI-powered insights (via Gemini 2.5 Flash), and optionally scan receipts (OCR).
File structure
• Top-level:
• README.md, firestore.* files → Firebase rules & indexes
• frontend/ → all React Native code
• frontend/ hierarchy shows:
• App.tsx, navigation/ → app entry & screen routing
• components/ → UI pieces (ExpenseCard, etc.)
• src/contexts → React Context for Auth
• src/lib/firebase & src/lib/supabase → wrappers for data/auth backends
• src/screens → feature screens (Dashboard, AddExpense, Profile, etc.)
• src/services → business logic API calls (transactionService, userService)
• src/utils → formatting helpers
• No custom Node/Express folder—backend is serverless (Firestore or Supabase) plus third-party AI/OCR.
README for additional insights
• Confirms tech stack: React Native, TypeScript, Firebase/Node + Express (but code shows serverless Firebase or Supabase), AI via Gemini REST API, OCR via Cloud Vision or Tesseract.
• Features/auth flow, charting, prediction, and prompt patterns.
Final guidance
• Break the system into as many discrete pieces as possible (UI screens, state layer, API wrappers, external services).
• Data flows end-to-end: from user tapping “Add Expense” to Firestore write, to AI prompt, to visualization.
• Be very detailed: annotate the transport mechanisms (HTTPS, WebSocket), authentication tokens, error handling paths.
• Iterate with the team to refine details (e.g., exact Firestore collections, Supabase functions, caching strategies).