Rupee-Wise

Dashboard

Inspiration

Managing finances in Indian households is often chaotic. Expenses are split across multiple family members, payments happen through UPI apps, cash, bank transfers, and receipts are rarely organized. Most finance apps available today feel too generic and are not designed around Indian spending habits.

The idea for RupeeWise came from observing how difficult it is for families to answer simple questions like:

“Where did most of our money go this month?” “Why are expenses suddenly increasing?” “How much are we spending on food delivery apps like Swiggy and Zomato?”

We wanted to build an intelligent finance assistant that feels less like a spreadsheet and more like a smart companion for Indian households.

🚀 What RupeeWise Does

RupeeWise is an AI-powered personal and family finance platform designed specifically for India.

Users can:

Upload receipts and bank statements Automatically extract transaction details using OCR Track expenses across family members Get AI-generated spending insights Detect unusual spending patterns in real time Receive personalized budget suggestions

The chatbot uses RAG (Retrieval-Augmented Generation) to answer queries based on the user’s own financial data, such as:

“How much did I spend on groceries this month?”

instead of providing generic responses.

🛠️ How We Built It

We built the frontend using:

React + TypeScript for a fast and scalable UI Recharts for analytics visualizations Modern responsive design inspired by fintech applications

For the backend:

Supabase handled authentication, PostgreSQL database storage, and real-time updates Google Gemini AI powered the intelligent chatbot and financial insights OCR.space API extracted text from receipts and statements Real-time anomaly detection pipelines monitored suspicious or unusual spending activity

We also implemented category-based spending analysis for Indian platforms like:

Swiggy Zomato DMart Blinkit Amazon India 📊 Intelligent Analytics

The system analyzes user spending patterns mathematically to identify anomalies and trends.

For example, we compare normal spending averages against current expenses:

Anomaly Score= Standard Deviation Current Spending−Average Spending

If the anomaly score crosses a threshold, users receive instant alerts about unusual expenses.

This helps families identify overspending before it becomes a serious issue.

⚡ Challenges We Faced

One of the biggest challenges was handling messy financial data.

Receipts from different stores had:

different formats unclear text inconsistent item names mixed English and regional language content

OCR extraction was not always accurate, so we had to build fallback cleaning and categorization systems.

Another major challenge was making AI responses accurate and contextual. Instead of generating random financial advice, we used a RAG pipeline so the chatbot responds only using the user's real transaction history.

Real-time syncing across multiple family accounts was also difficult because we needed updates to reflect instantly without slowing down the application.

📚 What We Learned

Through this project, we learned:

How RAG systems work in real-world applications Building scalable real-time systems with Supabase Integrating AI with structured financial data OCR processing and document parsing Designing user-friendly fintech dashboards Handling authentication and secure data flow

Most importantly, we learned that great AI products are not just about advanced models — they are about solving real problems for real people.

🌍 Impact

RupeeWise aims to help Indian families:

Understand spending habits clearly Prevent unnecessary expenses Build healthier financial discipline Save up to 15–20% monthly through actionable insights and anomaly detection

Instead of simply tracking expenses, RupeeWise helps users make smarter financial decisions every day.

🔮 Future Plans

We plan to add:

UPI integration Voice-based multilingual AI assistant Predictive monthly expense forecasting Investment and savings recommendations Credit score monitoring AI-generated financial health reports

Our vision is to create a complete AI-powered financial ecosystem built specifically for India.