Catchup - Personal Finance Application

Main screen
Create category
Import transactions
Date picker
Category view with graph
Category recommendation
Slide delete and range dropdown
Pick date range
Create transaction
Delete confirmation

Inspiration

As freshmen college students, we have had to take on a lot of responsibilities as we transition into this new phase of our lives. In this transformative period, we recognize the importance of tracking expenses to ensure long-term financial stability. With the currently available options on the market, we decided to make a custom app that helps automate our financial goals.

What it does

Our app allows users to create categories for expenses they want to track. Users can set a limit for each category to get a visual representation of their category goals. Within each category, users can view their transaction history through a bar graph. While users can add their transactions themselves, they can upload their Paypal or Chase account bank statements and our app will be able to parse the same and add individual transactions to the app. The user can then click on individual transactions and assign them to the relevant category. We have also incorporated an AI model to predict the category for a transaction based on its details.

How we built it

We decided that a mobile application would be the best medium for our project idea. We decided to use Flutter due to its cross-platform functionality. Further, we decided to use SQLite for its privacy and security features. To build the category recommendation system, we used Cloudflare's AI model. Finally, python to build parsers for the account statement files.

Challenges we ran into

Using different frameworks on languages for different functionalities presented a massive problem. While we were able to run the individual functions properly, putting them all together in our Flutter app proved to be a task by itself. We were able to fix these issues by setting up external back-end servers to act as an information exchange portal between Flutter and the other languages. Otherwise, we had a major challenge in improving the functionality of Cloudflare’s model. We had to spend a great deal of time on prompt engineering to improve the accuracy of our model.

Accomplishments that we're proud of

This was the first time using Flutter for all of us. We are incredibly proud of building a fully functioning application given the time constraints of a hackathon. In the spirit of learning things for the first time, it was a great accomplishment to be able to implement an AI model in our app.

What we learned

This was the first time we collaborated on a programming project. We had to effectively use version control systems and communicate to maintain our pace for the project. Also, As we mentioned above, we learned how to put together multiple technologies together to put together one functioning application.

What's next for Catchup

Currently, Catchup can only parse data from Chase bank statements and Paypal. In future versions of our app, we will be expanding access to more banks and financial institutions. Further, we also wish the remove the need for the user to upload their bank statements themselves. Instead, our app will be able to directly connect with the user’s financial institution and populate the app with their transactions. Finally, we plan on allowing users to make their own accounts. This will allow the user to synchronize their data through multiple devices.