Inspiration

Have you ever gone out with friends, enjoyed a meal, travelled together, or planned a trip, only to find that splitting the expenses becomes more complicated than the experience itself?

We've all experienced situations where someone pays the bill, someone forgets their share, and everyone starts calculating who owes whom. Sometimes, even a simple expense turns into an unnecessarily confusing process.

That everyday problem inspired me to create SplitMate AI.

I wanted to build something that would make sharing expenses easier, smarter, and more transparent. Instead of spending time calculating amounts, checking bills, and reminding friends about payments, people should be able to focus on enjoying their time together.

What it does

SplitMate AI is an AI-powered expense-sharing platform designed to simplify the way people manage shared expenses.

The idea is simple: upload a bill, let AI understand it, divide the expenses among group members, and make payments easier through PayPal.

Using OCR and Google's Gemini AI, SplitMate AI can extract important information from uploaded bills, identify expense categories, and help users organize their shared spending.

Our platform also calculates individual shares accurately and integrates PayPal Sandbox to demonstrate a convenient payment experience.

We want to turn a traditionally frustrating task into a smooth and intelligent experience.

How we built it

We chose technologies that would help us create a practical, functional application rather than just an attractive interface.

Python and FastAPI: To build the backend and handle application logic.

HTML, CSS and JavaScript: To create an interactive and user-friendly frontend.

Google Gemini AI: To understand bill information and assist with expense categorization.

Tesseract OCR: To extract text from uploaded bill images.

SQLite: To store group information, expenses and payment records.

PayPal Sandbox: To demonstrate secure payment processing in a test environment.

One important decision was to ensure that AI assists with understanding expenses, while the backend handles and validates the actual financial calculations. This helps make the application more reliable.

Challenges we ran into

One of our biggest challenges was figuring out how to make bill understanding reliable. Bills can have different layouts, formats, and item descriptions, making information extraction difficult.

Another challenge was ensuring that expense calculations remained accurate, even when multiple people shared different items or contributed different amounts.

Integrating AI with backend services and connecting PayPal's payment workflow also required careful planning.

We approached these challenges by separating the application into smaller modules, validating extracted information, keeping financial calculations within the backend, and testing individual components before connecting everything together.

Accomplishments that we're proud of

Our goal is simple: Less calculating, less confusion, more transparency, and better experiences together.

What we learned

Building SplitMate AI helped us understand how different technologies can work together to solve a real-world problem.

We explored AI integration, image-to-text processing, backend development, API communication, database management and payment integration.

More importantly, we learned that building a useful AI application is not simply about connecting an AI model. It is about identifying a genuine problem, choosing the right technologies, and creating a solution that people can actually use.

Share this project:

Updates

Submission history