LeetQuest

Inspiration

How we as CS students often have a hard time keeping a consistent schedule doing Leetcode

What it does

The project models a choose your own adventure game that is progressed forward by solving Leetcode problems. When a user enters the application, they are presented with the genres that they can tackle to practice. Upon clicking one, they have a unique story generated for them along with an accompanying decision tree, and the 'path' that they choose to go down will be blocked by a leetcode problem that they need to solve. Once they solve it, they can move on further, but the success of their traversal through these different paths will be affected by how well or optimally a user solves these leetcode problems. This continues until they reach the end of 5 rounds or the user finishes their practice for the day.

How we built it

We used a flask backend, that uses leetscrape to get the entire Leetcode questionbank. We then used a React frontend, and used flask to connect the data from the leetscrape dependency to the frontend. We then also implemented, using Open AI, a chat bot named Bella to give us live feedback on our code. If we need any help, Bella can provide us with hints or even feedback!

Challenges we ran into

We had a hard time figuring out the Open AI chatbot. We eventually got it to work, but kept on running into token errors :( Another challenge we ran into was trying to link all of the files together, as we all worked on separate parts of the code. However, we thought that this was a great team building exercise! We couldnt get backend in vercel :(

Accomplishments that we're proud of

Having never met before, we are proud of how we came together as a team to finish (sort of) a web app using a diverse tech stack. Having never worked with Open AI, we were keen to learn and read documentation! We are proud of the progress we made in the short time, and we are proud of how much we learned! We better programmers today than the Friday before DivHacks, and we are extremely proud of that!

What we learned

We learned a lot about AI, how to use Open AI in our projects, and we learned how to link a flask backend to a React frontend. Although we only had 30ish minutes together, the experience and knowledge we gained are very valuable, as we experienced working on code with others, as well as implementing tech stacks we have never used or have little experience with!

What's next for LeetQuest

We hope to improve on the generation of stories and responses from Bella or all of our generative AI models in general. They performed their tasks, but there was more fine tuning that we could have done that would have taken the context, narrative, and impact of the stories (and thereby the leetcode lessons) to the next level. Not only that, but Bella could have had integration with text to speech technologies that would've taken user experience up several notches as well.