Inspiration

Money is part of everyone's daily life, but understanding it isn't always easy. We often find ourselves asking questions like, "Where did my money go?", "How much do I actually have?", or "Can I afford this?"

We wanted to change the way people interact with their finances. Instead of making users navigate through multiple screens and transaction lists, what if they could simply talk to their money?

That idea became Ellie. We envisioned an AI-powered financial companion that makes managing money feel less intimidating, more accessible, and more human.

What it does

Ellie transforms the PayPal experience into something more conversational and intelligent. Users can ask questions about their financial activity and receive contextual answers based on their actual PayPal data.

Ellie is designed to help users:

  • Understand their PayPal balance and wallet information.
  • Review and understand transaction history.
  • Identify payments, incoming money, and outgoing expenses.
  • Ask financial questions through natural conversation.
  • Access supported PayPal wallet actions.
  • Interact with their assistant more conveniently through Bubble Mode.

Rather than simply displaying numbers, Ellie aims to help users understand what those numbers mean and make more informed financial decisions.

How we built it

We brought together mobile development, backend integration, and AI to create Ellie.

Our Android application serves as the primary user interface, while our backend connects the application to PayPal services and handles the communication between the different components.

We integrated an AI-powered conversational system using Groq, allowing Ellie to interpret user questions and generate responses using relevant financial information retrieved from the connected PayPal data.

We also developed features such as Bubble Mode and wallet interactions to make the experience more accessible and practical.

Our focus was not just to build another chatbot, but to connect AI with real financial functionality.

Challenges we ran into

We started with only three days to turn our idea into something real.

Three days to plan, develop, integrate, troubleshoot, and bring multiple technologies together. There were moments when things didn't work as expected, especially when connecting the AI, backend, and PayPal functionality.

Making an AI understand financial context was also a challenge. It wasn't enough for Ellie to generate convincing answers. She needed access to relevant data and had to respond based on actual information rather than assumptions.

Time was another major obstacle. We had more ideas than we could implement within the deadline, so we had to prioritize, make compromises, and keep moving forward.

We didn't have everything figured out from the beginning. We simply kept building, solving problems one at a time, and doing our best to bring Ellie to life.

Accomplishments that we're proud of

Our biggest accomplishment is seeing Ellie become more than just an idea.

What started as a concept eventually became a working application bringing together AI, PayPal integration, financial data, and mobile interaction.

We're especially proud that we managed to bring these different components together within just three days. More importantly, we created something that we genuinely believe has room to grow beyond the hackathon.

Finishing Ellie reminded us that meaningful projects don't always begin with perfect plans or unlimited time. Sometimes, they begin with an idea and a team willing to make it happen.

What we learned

This project taught us far more than we initially expected.

We gained a deeper understanding of AI integration, prompt engineering, backend communication, API handling, and connecting intelligent systems with real-world applications.

We also learned more about financial data, currencies, transaction structures, and the importance of accuracy when working with people's money.

Most importantly, we learned that building an AI application involves much more than connecting a model to a chatbot. The AI needs context, reliable data, proper system architecture, and carefully designed interactions to become genuinely useful.

We learned how to work under pressure, adapt when things break, collaborate, and turn limitations into opportunities to improve.

What's next for Ellie

For us, Ellie is not the end of a hackathon project. It's the beginning of something much bigger.

We want to continue developing Ellie into a more capable, intelligent, and personalized financial companion. We envision better financial insights, smarter spending analysis, improved contextual memory, broader currency awareness, and more seamless PayPal interactions.

We also want to improve its reliability, security, accessibility, and overall user experience.

Our ambition is to make Ellie something people can genuinely rely on, not just during a demonstration, but in their everyday financial lives.

We started with three days, one idea, and a team determined to make it happen.

We hope this is only the beginning of Ellie's story.

Built With

Share this project:

Updates