Rinder

Rinder Analyzing
Result Page

Inspiration

People are lonely, especially in this modern internet world. So many of us are on Reddit, why not find a soulmate through it? It is a bit like a marriage pack.

What it does

You type in your Reddit username, wait for a bit, then it will give you top three most similar users to you on Reddit

How we built it

We scrape from the Reddit API, passed on the information to Google Gemini, which will do a similarity analysis on the 15 most recent activity users on the top topic you browse. The top 3 and the reason why you match will be displayed on the interface

Challenges we ran into

We tried to build an Expo app, got everything in the front end set up, but eventually did not figure out how to connect it to the back end. We settled for a Flask app instead. We initially wanted users to sign into their reddit account, do the authentication, and access more data, but that was much more work and also make it hard to people to try it. We eventually decided to do the username input.

Accomplishments that we're proud of

Resolving web scraping, using the Reddit API Learning new things: first time working with Gemini and Flask

What we learned

It is important to start from basic things before expanding. Do not make the connection between front end and back end at the very end.

What's next for Rinder

Find an easier way to do authentication to have access to more data (upvotes downvotes). Optimize the AI to make it run faster. Better layout. Restore the Expo mobile app that we initially intended.