InvoTrack

Smart inventory & financial management system. Automates stock updates from delivery notes via GenAI, provides actionable business insights, and syncs real-time with POS.

A full-stack SaaS platform built with Next.js 16, Firebase, and Google Genkit. It bridges the gap between physical paperwork and digital business management by automating document processing and synchronizing inventory across systems.

🌐 Live Demo

You can try the live application here: invo-track-seven.vercel.app

InvoTrack.mp4

💡 How It Works

InvoTrack creates a seamless automated workflow for small businesses:

1. AI Document Agents: Uses Google Genkit to parse Hebrew invoices and delivery notes with high precision, extracting line items, prices, and supplier details.
2. Automated Inventory Control: Scanned delivery notes immediately update stock levels in the system, eliminating manual data entry errors.
3. Bi-Directional POS Sync: Real-time synchronization with external POS systems (Caspit, Hashavshevet) ensures stock levels and prices are always accurate across all channels.
4. 360° Business Intelligence: Aggregates data into a visual dashboard enabling tracking of expenses, supplier payments, and real-time profitability (KPIs).

✨ Key Features

• Generative AI Parsing - Extracts structured data from images (Invoices/Delivery Notes) using Gemini AI models.
• Smart Inventory Management - Automatic stock adjustments based on document scans and low-inventory alerts.
• POS Integrations - Plugin-based architecture supporting sync with Caspit and Hashavshevet.
• Financial Control - Track expenses, manage supplier payments ("Paid"/"Pending"), and monitor cash flow.
• Analytics Dashboard - Real-time graphs for monthly expenses, revenue, and top-selling items.
• Barcode Scanner - Built-in scanner for quick product lookup and management.
• Localization - Full support for Hebrew (RTL) and English currencies and date formats.

🛠️ Tech Stack

Frontend:

• Next.js 16 (App Router & Server Actions)
• React 19
• TypeScript
• Tailwind CSS & Shadcn UI
• Recharts (Analytics)

Backend & AI:

• Google Genkit (AI Agent Framework)
• Google Gemini (LLM)
• Firebase (Firestore, Auth, Storage)
• Node.js Server Actions

Architecture:

• Adapter Pattern - For scalable POS system integrations.
• Server Actions - For type-safe backend logic execution.
• React Query - For optimized server state management.

📁 Project Structure

src/
├── app/              # Next.js 16 App Router pages
├── ai/               # Genkit flows and prompt engineering
├── actions/          # Server Actions (Backend logic)
├── services/         # Business logic & POS Adapters
├── components/       # Reusable UI components (Shadcn)
├── hooks/            # Custom React hooks
├── lib/              # Utility functions & Firebase config
└── locales/          # i18n translation files

🎯 Key Technical Highlights

• Cutting-Edge Stack: Built on the latest Next.js 16 and React 19, utilizing modern features like Server Components and Actions.
• Robust AI Engineering: Implements retry mechanisms and exponential backoff for AI flows to ensure reliability in production.
• Scalable Architecture: The POS integration layer uses the Factory and Adapter patterns, making it easy to add new POS providers without changing core logic.
• Type Safety: End-to-end type safety with TypeScript and Zod schema validation for AI outputs.

📄 License

This project is MIT licensed.

👤 Author

Omry - Full-Stack Developer

Feel free to reach out for collaboration or opportunities!

---

Built with Next.js, TypeScript, and Firebase