Wealth & Health

The login/register page
The dashboard page
The page to save new records
The maps page to see the average cost of an area
Summary of average cost for a certain area
Output with spending advice
Output with spending prediction
The settings page

Inspiration

In today's fast-paced world, we often overlook on how our spending habits impact our wellbeing. The average person tracks their finances and health separately but are missing on crucial connections between the two. Our team noticed this and wondered, what if an app could help us visualize and give advice on these relationships.

What it does

Wealth and Health is a mobile application that tracks your expenses as well as how the spending affects your wellbeing. By categorizing spending across areas like groceries, dining out, healthcare, entertainment and etc., our app utilizes AI analysis to provide personalized insights about your expenses and also how your financial decisions affect your overall wellbeing. The app offers a user-friendly dashboard, expense entry page, location-based spending maps, and AI-generated recommendations that suggests healthier financial choices.

How we built it

We developed a Flutter frontend with a responsive design. The backend uses Express.js with AstraDB (DataStax) for our primary database and Gemini API to provide a budget estimate as well as spending advices based on the spending categories. We also integrated Google Maps API for location-based analysis.

Challenges we ran into

Integrating Langflow was our biggest challenge and unfortunately we couldn't resolve all the implementation issues within the hackathon and went with an alternate approach where we make direct calls to the AI agent from our backend.

Accomplishments that we're proud of

We successfully implemented a vector-based recommendation system that improves as users add more data, creating increasingly personalized insights. Our location visualization feature maps spending patterns to locations, revealing geographic trends. We're proud of our spending advice and prediction that provides insights even without the health advices.

What we learned

Building this project taught us about the technical challenges of vector databases and embeddings. We learnt about a new database Astra DB, and creating AI agents using LangFlow, even though, we weren't able to configure it to our satisfaction due to time restrictions.

What's next for Wealth & Health

We plan to fix and enhance the health analysis component by properly integrating Langflow and refining our AI prompt for more meaningful health insights.