Spending Ghost

Inspiration

I wanted to make budgeting more insightful by directly showing consumers how their everyday purchases could have grown if they invested them instead. People often regret wasteful spending, so Spending Ghost turns that regret into a useful reflection.

What it does

Spending Ghost lets users track their purchases and shows the “ghost portfolio” they could have had if that money had been invested in the merchant’s stock. It uses local time zones and real-time stock data to give insights as both an integrated portfolio and for individual merchants.

How we built it

I built the backend classes, including a class for transactions, merchants, and portfolios, in Python with Flask. I also used SQLite for database storage, yfinance to fetch historical and real-time stock data, and datetime and tzlocal for dates, times, and time zones. The app is hosted on Render and linked to a custom domain with a clean HTML interface for the frontend.

Challenges we ran into

I had to figure out how to accurately fetch stock prices at specific timestamps, including when markets were closed, which was difficult because I had to experiment with the yfinance library and understand all its documentation. Connecting the backend logic to a clean front end and deploying it without issues also took some trial and error because I didn't have a lot of experience with frontend development and hosting websites.

Accomplishments that we're proud of

I'm proud of turning a unique idea into a fully functional, live web app with minimal resources. This is my first ever hackathon, and I knew developing a well-rounded app with both a backend and a frontend on my own would be difficult, so I was happy that it all came together nicely in such a short time. The app smoothly integrates user-specific storage with databases, real-time stock data, and a clean, usable interface.

What we learned

I learned how to deal with stock APIs, time zone conversions, Flask app structure, some basic HTML, SQLite databases, and deployment of a web application to a custom domain. I also gained some experience designing around real-world scenarios like changing market hours and stock prices.

What's next for Spending Ghost

I would like to add AI-driven analysis that suggests which habits are costing the most in the long term to help more with budgeting. Also, I want to improve the UI because it's bland, but I need to learn more frontend development for that. Lastly, I want to support more investment options besides just stocks.

Built With

• flask
• html
• python
• render
• sqlite
• yahoo-finance