Inspiration:

Being a student in the age of AI is difficult, a study recorded that 86% of students use AI in their studies but how many of them use AI properly is concerning, it was recorded that nearly 83% students who use AI in their studies can't even recall what they generated. The reason for this is they use AI as a dependency, which decreases both attention span and effort. Our team wanted to change that, which is why we created ArcadiaU.

What it does:

Our app gamifies learning, taking professors lecture materials and then transforming them into mini games that track progress and growth, to make learning more engaging and interactive. With several learning module games based off of a variety of subjects, our app uses ai as a tool, creating questions from lectures for practice and giving access to a voice AI teaching assistant trained on ElevenLabs to be used for guidance not solely for answers. Additionally, users can review their missed/wrong questions in the office hours tab where they also have access to the AI Teaching Assistant. Additionally, we setup a login page utilizing Solana's blockchain to authenticate users. When a user enters their username, the key-pair associated with the username is retrieved from MongoDB, and a challenge string is created and is signed with the username's private key. That signature is verified with a wallet id or public key from MongoDB and if successful which retrieves the user's information regarding stats, modules, etc. Our app is integrated both as desktop application and a mobile application.

How we built it

The backend is built on python where it builds the Solana-based authorization, lecture parsering, AI question generation using Gemini API, and using MongoDB to hold all data.

The frontend consists of the UI and the Flappy Bird game that's all built in HTML, CSS, and JavaScript. The Flappy Bird game is built on the Canvas API. The dashboard and login page is also handled here.

The bridge between them is Rust and Tauri. Tauri wraps the web frontend into a native desktop window and provides a secure stdin/stdout bridge that lets JavaScript invoke Python commands in real time.

AI Teaching Assistant agent trained and created on ElevenLabs, and integrated in app using frontend languages such as javascript.

Challenges we ran into

-Figuring out how to use MongoDB in python -Integrating authentication for the login page with Solana and MongoDB -Issues with the game logic -Figuring out how tauri works

Accomplishments that we're proud of

  • Got our app fully up and deployed, with all core functionalities intact
  • Was able to stay up throughout the night with energy and motivation
  • Being able to complete a 24 hr hackathon with new and returning teammates

What we learned

  • we learned a lot for some of us it was our first hackathon for others it wasn't, but regardless of that we learned an immense amount, and had a great experience. Our team acquired new skills such as Tauri, rust, and using the Solan Blockchain additionally for most of us it was our first time developing an application, and greatest of all we made new connections with industry professionals and like minded peers.

What's next for ArcadiaU

  • We want to get feedback from students by deploying it within a university test environment to further improve the app
  • Further develop the User interface
  • Add a larger selection of games, for the learning modules

Built With

Share this project:

Updates