Inspiration

We built Spendwise around a common shopping frustration: finding a suitable product often means searching across multiple stores, comparing options, and trying to stick to a budget. As a team, we wanted to explore how an AI shopping assistant could make product discovery more focused and help people shop according to their own spending limits.

What it does

Spendwise lets shoppers search for products with an AI assistant and set spending policies such as a maximum price per item or per order. It also includes Google sign-in and a PayPal Sandbox subscription checkout for demonstrating the account and payment flows.

How we built it

We built the frontend with React and the backend with Spring Boot, using PostgreSQL for data storage. The backend connects the app’s search, policy, authentication, and payment features. We deployed the frontend on Vercel and the API on Render.

Challenges we ran into

Connecting the separate parts of the system took careful debugging. We worked through authentication and deployment configuration, API proxy errors, and the difference between approving a PayPal Sandbox payment and completing a merchant-side checkout. Render’s free-tier cold starts also made the backend temporarily unavailable while it woke up.

Accomplishments that we're proud of

As a team, we brought together product search, personalized spending limits, sign-in, and a payment demo in one full-stack application. We’re proud that Spendwise is deployed and that we learned to diagnose issues across both the application and its hosting services.

What we learned

We learned how frontend and backend services communicate, how authentication and database migrations fit into a full-stack project, and how to approach payment integrations in a sandbox environment. We also learned that deployment settings, proxy configuration, and hosting behavior can affect the user experience just as much as application code.

What's next for Spendwise

We’d like to improve product comparisons, make search recommendations more useful, and strengthen the checkout experience. We also want to reduce backend startup delays and test the complete payment flow more thoroughly before considering a production payment setup.

Built With

Share this project:

Updates

Submission history