Inspiration

We’re college students, we spend too much money, and somehow we never know where it went. We wanted to build something that doesn’t just track spending, but actually holds us accountable before we make the same mistakes again.

What it does

WhereIsMyMoney lets users log purchases by talking or snapping a picture of a receipt. It automatically identifies the amount, merchant, and category. Users set weekly budgets for categories like Food or Shopping, and if they overspend, the app actually calls them out—literally. At the end of the week, it also calls with a spending summary.

How we built it

We used ElevenLabs for voice interaction and NVIDIA Nemotron to understand transactions and extract information like price, merchant, and spending category. We connected these systems to our backend to store transactions, track budgets, and trigger automated calls when necessary.

Challenges we ran into

Our biggest challenge was getting all the different parts of the app to communicate reliably. Turning messy, natural user input into structured financial data and correctly categorizing purchases also required a lot of testing.

Accomplishments that we're proud of

We’re proud that we built more than another expense tracker. WhereIsMyMoney actively interacts with users and responds to their spending behavior. We were also able to combine voice, AI, receipt processing, budgeting, and automated calls into one working application.

What we learned

We learned a lot about integrating different APIs and AI models into a single application. We also learned how difficult it can be to turn unstructured information, like speech and receipts, into clean and useful data.

What's next for WhereIsMyMoney

Next, we want to connect directly to users’ bank accounts so transactions can be tracked automatically. We’d also like to add smarter spending insights, personalized budget recommendations, better analytics, and more ways for the app to intervene before users make purchases they might regret.