-
-
Login Page for MoneyMind
Inspiration
One of the major challenges I regularly face is tracking my spending, which occurs through various channels like cash, cards, bank transfers, etc. The only consolidated data regarding my income and expenses comes through my bank statements, as such, I wanted to have my own analysis on the data as well as try to optimize my spending habits to achieve my set goals and pay off the debt for my education loan.
What it does
MoneyMind is an AI finance analyst that works with your actual data. You start by uploading a bank statement. The app extracts and lists all your transactions in a clean, editable table. You can then set your financial goals and log any debts in separate tabs. The core feature is the "AI Plan" button: with one click, Gemini 3 analyzes all your uploaded transactions, goals, and debts together. It then provides a personalized report detailing your spending habits, pinpoints areas to cut costs, and offers specific steps for better financial planning tailored just for you.
How we built it
We built MoneyMind as a modern web application using React and TypeScript for a responsive front-end, with Google AI Studios as our build tool. The backend logic and AI integration are handled server-side. The core of the project is the Gemini 3 API. We used its advanced document understanding capabilities to parse text from uploaded bank statements and its powerful reasoning to perform contextual analysis of the combined financial data (transactions, goals, debts) to generate coherent, personalized advice. The entire project is uploaded to GitHub and hosted using Netlify.
Challenges we ran into
One of the main challenges was structuring the unstructured data from bank statements, which can come in many formats, into a consistent transaction model for the AI to analyze. Another significant challenge was designing prompts for Gemini 3 that would reliably produce a structured, comprehensive financial analysis from a user's unique dataset, ensuring the advice was both practical and relevant.
Accomplishments that we're proud of
I am most proud of creating a seamless end to end workflow from document upload to actionable AI insight that feels intuitive. Successfully integrating Gemini 3 to act as a competent financial analyst on custom user data is a key achievement. I'm also proud of building a clean, user friendly interface that makes a potentially stressful topic (personal finance) feel manageable.
What we learned
I gained hands on experience with the Gemini 3 API, particularly in leveraging its document processing and complex reasoning for a specialized domain. I learned the importance of carefully structuring data and crafting precise prompts to get consistent, high quality outputs from a large language model. The project also deepened our understanding of user experience design for data-intensive applications.
What's next for Money Mind
I'm plan to add automatic transaction categorization and richer data visualizations like spending charts. I'm also aiming to support connections to bank APIs for automatic data syncing. Expanding the AI's capabilities to offer interactive Q&A about the generated plan and developing a mobile app are key steps to make MoneyMind an even more powerful daily financial companion.
Log in or sign up for Devpost to join the conversation.