Inspiration
Going through college we always wanted a platform that could teach us better and a platform that could help us since we each have different learning styles and with the power of AI personalization we believe this platform can help other students worldwide learn better. We wanted to create something that students would enjoy using—something that makes studying together feel less like a chore and more like a game. The idea was to help learners get smarter and have fun while doing it. We were also inspired by the possibility of submitting this project to multiple hackathon tracks by combining web development, education, and AI in a meaningful way. Our goal was to design a tool that supports intuitive user learning and provides helpful feedback tailored to each individual.
What it does
Our web app turns studying into a multiplayer game powered by AI. Players choose individual subjects that each of them wants to focus on, and the app uses Gemini AI to generate personalized questions. Players take turns answering questions—if they get it right, the screen flashes green and they damage their opponent; if wrong, it flashes red and they damage themselves. The players battle each other until one player is out of health or they've gone through all the questions. Behind the scenes, the app tracks which questions each player struggles with. At the end, it provides a tailored review that tells the players what they should work on.
How we built it
We built the app using HTML, JavaScript, and CSS for the front end. The AI integration was done using Gemini’s API, which generates topic/subject-based questions in real-time. We process and clean the AI responses into a JSON structure, which the front end reads to display questions to the players. Game logic includes turn-taking, answer evaluation, and visual feedback. We also implemented a system that analyzes each player’s incorrect answers to generate a personalized learning path using additional AI prompts.
Challenges we ran into
One of our biggest challenges was coordinating between different programming languages and frameworks. We had to deal with integrating the AI (backend in python) with the front-end (Javascript) logic, which led to errors and difficulties in passing data cleanly across components. Managing the structure of dynamically generated content and keeping the game state consistent for multiple players was also tricky.
Accomplishments that we're proud of
We’re proud of building a functional multiplayer learning experience in such a short time, especially with real-time AI integration. We managed to combine fun and education while also offering personalized feedback. It was exciting to see how AI can be used and adopted in education to enable students like us to learn better and smarter. It was fun to do this in a gamified version as this boosted the spirit to learn in a fun way.
What we learned
We learned how to integrate AI into a web application, manage data formats (like stripping JSON to fit front-end needs), and deal with communication across the stack using APIs and wrapping one API with another. We also got hands-on experience building a game loop, tracking player progress, and designing user-friendly feedback systems. Providing feedback based on what was failed. Lastly but importantly, we also learnt to think in terms of what we can add to personalize the application like adding audio elements to our application for people who prefer audio since education should be accessed by all and also how to teach the ai model to learn areas and the techniques a user does when they fail so the model can help them better .
What's next for ELearn
We want to use the power of AI (Gemini AI) to study common patterns in how users struggle with certain topics and personalize their learning experience. By identifying specific areas and techniques a user finds challenging, the system can adopt an instructional plan tailored to their needs helping users learn how to succeed by focusing on how they fail to create a better structured lesson plan for each user. We strongly believe this will eventually help students like us to become smarter even while we struggle with tough subjects. This will help us understand the content better by using adoptive techniques. Another goal we will do is to provide deeper analytics for players, such as performance graphs to monitor how they learn the content. We also plan to polish the multiplayer experience and add more subjects, question types, and even voice support for accessibility. Eventually, we want ELearn to be a go-to platform for gamified education that adapts to each learner and can improve education thus helping users.
Log in or sign up for Devpost to join the conversation.