AI-powered Personal Finance Intelligence Platform built with the MERN Stack.
FinSentry helps users effortlessly manage their finances by combining AI-powered receipt processing, transaction management, financial analytics, and automated monthly reports into a single modern workspace.
Unlike traditional expense trackers that require manual data entry, FinSentry leverages Google Gemini AI to intelligently extract, categorize, and organize expenses from receipt images while giving users complete control before importing them.
π Application: https://finsentry-eta.vercel.app
- JWT Authentication
- Protected Routes
- Persistent Login
- Secure API Access
- Add / Edit / Delete Transactions
- Bulk Delete
- Smart Search
- Advanced Filters
- Pagination
- Category Management
- Payment Methods
- Recurring Transactions
Upload a receipt image and let Gemini AI automatically
- Detect purchased items
- Extract prices
- Categorize expenses
- Generate editable transaction drafts
Supports receipts containing multiple expense categories.
Example
Milk βΉ80
Medicine βΉ250
Shirt βΉ799
β
Grocery βΉ80
Healthcare βΉ250
Shopping βΉ799
Every extracted transaction can be reviewed before saving.
Interactive dashboard featuring
- Total Income
- Total Expenses
- Balance
- Savings Rate
- Cash Flow Trends
- Expense Breakdown
- Recent Transactions
Bulk upload transactions using CSV files.
Generate AI-powered monthly financial summaries and send them directly to users via email.
Traditional expense trackers require users to manually enter every transaction, making personal finance management repetitive and time-consuming.
FinSentry simplifies this process through AI-powered automation while ensuring users remain in complete control of their financial data.
The project was built with a production-oriented backend architecture focusing on modularity, scalability, validation, and clean API design.
ββββββββββββββββββββββββ
β React Client β
β Dashboard & UI β
ββββββββββββ¬ββββββββββββ
β
Axios HTTP Client
β
βΌ
ββββββββββββββββββββββββ
β Express Server β
β REST API Layer β
ββββββββββββββββββββββββ€
β Authentication β
β Validation (Zod) β
β Controllers β
β Business Services β
ββββββββββββ¬ββββββββββββ
β
βββββββββββββββββ¬βββββββββ΄ββββββββββ¬βββββββββββββ
βΌ βΌ βΌ βΌ
MongoDB Gemini AI Cloudinary Node Cron
β β
ββββββββββββββββββββββββββββββββ¬βββββββββββββββββ
βΌ
Resend Email API
Receipt Image
β
βΌ
Cloudinary Upload
β
βΌ
Gemini AI Processing
β
βΌ
Extract Items & Prices
β
βΌ
Automatic Categorization
β
βΌ
Generate Transaction Drafts
β
βΌ
User Review & Approval
β
βΌ
MongoDB Storage
- JWT-based authentication
- Protected API routes
- Persistent sessions
- Zod request validation
- Centralized error handling
- Input sanitization
- Controller-Service pattern
- Modular route organization
- Reusable middleware
- Clean REST API design
- Secure image uploads
- Cloudinary integration
- AI-powered receipt parsing
- Monthly financial reports
- Automated email delivery using Node Cron + Resend
finsentry
β
βββ client
β βββ components
β βββ context
β βββ hooks
β βββ layouts
β βββ pages
β βββ routes
β βββ utils
β
βββ server
βββ config
βββ controllers
βββ middleware
βββ models
βββ routes
βββ services
βββ validators
βββ utils
- React
- Vite
- JavaScript
- Tailwind CSS
- React Router
- Axios
- Context API
- React Hot Toast
- Recharts
- Node.js
- Express.js
- MongoDB
- Mongoose
- JWT
- Zod
- Multer
- Node Cron
- Google Gemini AI
- Cloudinary
- Resend
git clone https://github.com/anupam-devcodes/finsentry.gitcd server
npm install
npm run devcd client
npm install
npm run devPORT=
MONGO_URI=
JWT_SECRET=
JWT_EXPIRES_IN=
CLOUDINARY_CLOUD_NAME=
CLOUDINARY_API_KEY=
CLOUDINARY_API_SECRET=
GEMINI_API_KEY=
GEMINI_MODEL=
RESEND_API_KEY=
RESEND_FROM_EMAIL=
CLIENT_URL=
The backend follows a RESTful architecture with dedicated modules for
- Authentication
- User Management
- Transactions
- Receipt Processing
- Analytics
- Reports
The API is organized using the Controller-Service pattern for improved maintainability and scalability.
- Avatar Upload
- Budget Planning
- Financial Goals
- Multi-Currency Support
- Spending Predictions
- AI Budget Recommendations
- Mobile PWA
- Dark Mode
- OCR Fallback for Low-Quality Receipts
Anupam Choubey
GitHub: https://github.com/anupam-devcodes
If you found this project helpful, consider giving it a β.