A full-stack receipt scanning application with AI-powered analysis. Upload receipts, extract items automatically, and track spending by category.
- AI Receipt Scanning: Upload receipt images, get items extracted automatically
- Category Management: Organize items with custom categories
- User Authentication: Register/login with JWT and per-user data isolation
- Multi-Currency Support: Track spending in EUR, GBP, USD with real-time conversion
- Analytics Dashboard: Monthly spending breakdown by category
- Advanced Filtering: Search and filter receipts by store, category, date range, and amount
- Export to CSV/PDF: Download filtered receipts as spreadsheet or PDF reports (with optional images)
- Dark Mode: Modern dark-first UI design
| Layer | Technology |
|---|---|
| Frontend | Next.js 16, React 19, TanStack Query, Tailwind CSS, shadcn/ui |
| Backend | FastAPI, SQLModel, Pydantic AI, PostgreSQL |
| AI | Google Gemini Vision |
# Clone and setup
git clone https://github.com/cgoncalves94/receipt-scanner.git
cd receipt-scanner
make setup
# Add your GEMINI_API_KEY and JWT_SECRET_KEY to backend/.env
# (see backend/.env.example for required vars)
# Start services
make dev # Terminal 1: Backend + DB
make dev-frontend # Terminal 2: Frontend- Frontend: http://localhost:3000
- Backend API: http://localhost:8000
- API Docs: http://localhost:8000/docs
See DEVELOPMENT.md for detailed setup, commands, and architecture guide.
See LICENSE file.