How code-ph0enix/finedge works
Project type and purpose
• FinEdge is a full-stack AI-driven personal finance management and investment advisory platform.
• Frontend: React + TypeScript application built with Vite and Tailwind CSS, rendering dashboards, onboarding flows, chat, and data-visualizations.
• Backend: Python Flask REST API implementing business logic, AI services (Gemini, OpenAI), data models, and integrations with external market/stock data and authentication providers.
File structure
Top-level separation:
• frontend/ – React UI code (components, pages, services, hooks, context)
• backend/ – Flask application, AI modules, data layer, utility scripts, backups, legacy code
Key observations in backend/:
• app.py – Flask application factory
• database.py & models.py – data layer
• ai_financial_advisor.py, recommendations.py, financial_journey.py, learning_progress.py – core business/AI logic
• /tools – internal helper scripts
• data-input.json, fin-path.json – sample data
• .env.example, requirements.txt – configuration
Frontend structure highlights:
• src/pages – high-level routes (Chatbot, Home, Learn, etc.)
• src/components – reusable UI pieces (ChatInput, DashboardLayout, Onboarding steps, etc.)
• src/services & src/utils – wrappers around HTTP calls (chatApi, stockApi, onboardingApi)
• src/context – global state (OnboardingContext, ThemeContext, TourContext)
README insights
• Clean architecture, Flask app factory, RESTful API design
• Unified AI service pattern (“FinancialAIService”)
• External integrations: Gemini AI, stock/market/ news APIs, OAuth providers
• Emphasis on separation of concerns and centralized error handling