ANT-icipa

ANT-icipa turns your daily small expenses into smart investments by automatically adding a chosen percentage to each purchase and investing it safely.

Comment

Inspiration

We were inspired by our own experiences, what we know were the problems us as students may have on our journey to financial responsibility. We wanted to create something that not only made us more accountable of our spending, but helped us build our future as well. Basically, what we thought was, ant expenses will happen here and there, so we should might as well use them as a tool to empower financial wellness (yes, as crazy as that sounds).

What it does

ANTicipa is a smart financial assistant that detects your ant expenses in real time and and automatically invests a percentage of those amounts into your capital one account. It connects to your transactions, identifies spending patterns, and classifies which ones are ant expenses using AI. Then, it shows you visual dashboards of your spending and savings and lets you chat with Capi, your AI financial assistant, who explains your spending habits and gives saving tips.

How we built it

We built ANTicipa using a combination of software development tools, cloud technologies, and machine learning frameworks. The mobile application was developed in Swift. For version control and project organization, we used GitHub to maintain a clean and collaborative workflow. On the backend, we implemented machine learning with Arm (TypeScript) to help Capi analyze spending behaviors and identify ant expenses. We integrated the Gemini API alongside Arm to deliver a personalized experience for each user, adapting recommendations based on their transaction history via a chatbot. Our data was stored and managed through a MySQL database hosted on Aiven, ensuring security and scalability. Additionally, Python was used for data processing and backend scripting, tying together our AI, analytics, and cloud components into a cohesive system.

Challenges we ran into

One of the biggest challenges we faced during development was creating our own database and API from scratch. Designing the database structure required careful planning to ensure scalability, efficiency, and security, especially as we needed to manage sensitive financial data and user-specific information. Integrating it with the API added another layer of difficulty, making sure that data flowed seamlessly between the app, the machine learning model, and the cloud storage without delays or inconsistencies. This process taught us how interconnected each part of the system is, and how critical a well-structured backend is to the overall performance and reliability of the app.

Accomplishments that we're proud of

One of our biggest accomplishments was developing a chatbot capable of providing a fully personalized experience for each user. By using the user’s unique ID upon login, the chatbot can access their transaction history to deliver personalized insights and recommendations. This feature not only makes the app feel more interactive and intuitive but also reinforces ANTicipa’s mission of turning small inconveniences and interactions into meaningful financial growth opportunities. Despite the difficulties we faced while creating our own database and API, we were able to overcome them through continuous testing.

What we learned

We learned how to build with Arm for machine learning, enabling Capi to identify when a purchase qualifies as an “ant expense.” For the app’s development, part of our team had to relearn Swift to create an efficient and user-friendly interface. We also implemented speech detection using the Gemini API, which enhanced the app’s interactivity and accessibility. Additionally, we learned how to manage and structure a cloud database, ensuring that our data was organized and scalable. Finally, we built our own database through the API and developed the ability to identify and filter out truly valuable data, which is essential for maintaining meaningful insights in our system.

What's next for ANT-icipa

We plan to connect ANTicipa directly to Capital One’s API, allowing users to link their accounts and automatically reinvest a chosen percentage of every “ant expense” into an investment account. Using machine learning, ANTicipa will continuously learn each user’s spending patterns to anticipate upcoming expenses, warn them before they overspend, and suggest smarter alternatives in real time. We aim to expand beyond iOS to create a multi-platform experience, empowering users to monitor, manage, and grow their finances anywhere, turning everyday spending into a habit of smart saving and investing.

Share this project:

Updates