🛠️ The WealthFam Tech Stack
WealthFam is built on a "Local-First, AI-Augmented" philosophy. Every technology in our stack is chosen for its performance, privacy characteristics, and developer experience.
🚀 Backend: FastAPI & Python 3.10+
We chose FastAPI as the central orchestrator for its modern features:
- High Performance: Built on Starlette and Pydantic, it is one of the fastest Python frameworks available.
- Type Safety: Leveraging Python type hints for automatic validation and interactive API documentation (Swagger/OpenAPI).
- Asynchronous Support: Perfect for handling multiple concurrent ingestion streams (SMS, Email, Web).
🦆 Database: DuckDB
WealthFam utilizes DuckDB for its primary data storage. DuckDB is an in-process SQL OLAP database management system.
- Local-First Privacy: Your financial data stays in a single file on your machine or container, never shared unless explicitly synced.
- Extreme Speed: Optimized for analytical queries (e.g., "Show me my spending by category over the last 12 months") with columnar execution.
- Zero Management: No server to install or maintain; it works as easily as SQLite but with modern performance.
🎨 Frontend: Vue 3 & Material 3
Our web interface is designed for clarity and responsiveness:
- Vue 3 (Composition API): Provides a clean, reactive architecture for complex financial dashboards.
- Vite: The next-generation frontend tool that ensures near-instant development starts and builds.
- Material Design 3 (M3): We strictly follow M3 principles to ensure the UI is accessible, beautiful, and feels native on every device.
- Tailwind CSS: For high-fidelity, utility-first styling that doesn't bloat the bundle.
📱 Mobile: Flutter
The WealthFam Mobile app (Android) is built with Flutter:
- Native Performance: Compiles to ARM machine code for smooth 60fps animations.
- Native Integration: Essential for our background SMS listeners and device hashing capabilities.
- Single Codebase: Allows us to maintain a consistent experience if we expand to iOS in the future.
🧠 AI: Google Gemini Pro
For unstructured data parsing (like messy credit card SMS or non-standard bank emails), we use Gemini Pro:
- Logical Reasoning: Gemini excels at extracting structured JSON from natural language financial alerts.
- Privacy-Conscious: We only send the minimum required text for parsing; no identifying PII is sent to the model.
| Layer | Technology | Role |
|---|---|---|
| Orchestration | FastAPI | API Gateways & Business Logic |
| Persistence | DuckDB | Precision Ledger & Analytics |
| Logic Engine | Python 3.10+ | Core Parsing & Merging |
| Interface | Vue.js 3 | Web Dashboard |
| Mobile | Flutter | Real-time SMS Ingestion |
| Intelligence | Gemini AI | Unstructured Text Parsing |
| Styling | Tailwind CSS | Material 3 Implementation |
*Engineering Excellence · Local Isolation · Precision Accuracy*