Pay Pals is a social financial wellness application that gamifies saving habits through a prediction market interface. By replacing monetary bets with social currency (favors and tokens), the platform facilitates group accountability and financial literacy for students and young adults without financial risk.
- Overview
- Features
- System Architecture
- Technologies Used
- Development Challenges
- Future Roadmap
- Contributors
Pay Pals was developed to address the lack of engaging financial tools for the younger demographic. The concept draws inspiration from prediction markets like Kalshi but pivots the core mechanic toward personal finance habits. Instead of wagering capital, users wager favors and SOL tokens on the likelihood of achieving shared financial goals generated by artificial intelligence.
- Secure Onboarding: Integration with Capital One data services and Solana wallet generation for secure user identity and asset management.
- AI-Driven User Profiling: Utilizes the Nessie API to analyze transaction history, with Google Gemini generating unique user Titles (e.g., "The Coffee Connoisseur") based on spending patterns.
- Lobby System: Allows users to create private groups and send Blinks (invites) to peers.
- Dynamic Goal Generation: Gemini analyzes the collective profiles within a lobby to generate context-aware shared financial goals.
- Prediction Market Interface: A social betting system where participants predict the success or failure of peers in achieving set goals.
- Incentive Mechanism: A dual-reward system utilizing on-chain tokens for winners and real-world social favors as settlement for lost wagers.
The application follows a decoupled client-server architecture.
- Framework: React (TypeScript) for a type-safe, component-based UI.
- Styling: Tailwind CSS for responsive design.
- Graphics: Integrated Three.js elements for 3D data visualization and modern UI aesthetics.
- Runtime: Python.
- Framework: FastAPI for high-performance, asynchronous API endpoints.
- Database: MongoDB Atlas for scalable, cloud-based document storage.
- Infrastructure: Deployed via Railway.
- Google Gemini 2.5 Pro: LLM integration for natural language processing, user profiling, and dynamic goal generation.
- Capital One Nessie API: Provides mock financial transaction data for simulation purposes.
- Solana API: Handles token logic and wallet management.
- Database Connectivity: Resolved TLS handshake errors when establishing secure connections between the Python application and MongoDB Atlas.
- Version Control: Managed complex merge conflicts arising from concurrent development on critical components.
- 3D Optimization: Optimized Three.js asset loading and rendering within the React virtual DOM to ensure performance stability.
- Deployment Configuration: Configured Railway for dual-service hosting to support both the static frontend and the FastAPI backend service.
- Microservices Orchestration: Successfully synchronized three distinct external APIs (Gemini, Nessie, Solana) into a cohesive user flow.
- User Interface: Delivered a polished, accessible interface through iterative design cycles.
- Team Workflow: Established a professional Git workflow for continuous integration and collaboration.