Inspiration

Our inspiration for this project came from the challenges I faced when preparing for exams. Often, students spend hours re-reading notes or using generic study apps that do not align with their own course material. We wanted to create an app that transforms a student’s own documents and lectures into interactive quizzes, making sure the study experience is personalized and effective.

We were also inspired by how there are already specialized tools for standardized exams like the MCAT, LSAT, and actuarial exams, which help candidates prepare with structured practice and adaptive learning. However, we noticed that there was not an equivalent tool available for everyday students preparing for a midterm, a final exam or even just trying to master weekly lecture content.

What it does

LevelUP allows users to upload their course documents and the tool automatically detects the topics. For each topic, it generates a short quiz. You repeat the quizzes until you master them and get a 100% in every one of them. Then, you proceed to take the final exam which serves as a cumulative exam which covers every topic, so you walk into the real exam already confident and prepared.

How we built it

We built LevelUP with using Next.js as the framework, used Gemini's model 2.5 flash for the AI generation and React.js for the frontend and TailwindCSS for styling.

Challenges we ran into

One of the biggest challenges was fine-tuning the prompts given to the language model to make sure quizzes were high quality and aligned with the uploaded content. We initially used GROK API, but the results were inconsistent. After iterating, refining prompt structures, and experimenting with different models, we achieved more reliable quiz generation using Gemini API.

Accomplishments that we're proud of

  • Built a working prototype that allows users to upload course documents and lecture recordings, which are then transformed into personalized quizzes
  • Successfully implemented a topic-by-topic mastery system, making sure students achieve 100% before progressing to the cumulative final exams
  • Created study tools usually reserved for big exams like the MCAT or LSAT into a format that everyday students can use, making structured and adaptive studying for accessible

What we learned

  • Strengthened our skills in full-stack development, balancing the frontend UI/UX with the backend logic and data flow
  • The importance of user progress tracking
  • Usage of APIs
  • Importance of teamwork, communication and collaboration

What's next for LevelUP

Future features we are hoping to implement are to save the progress, add multiple files for the same session, share the session to a friend and image generation, implement and tailor within current education platforms like D2L. We are also hoping to integrate a recording feature which lets users record their class lectures and lets the AI create quizzes and exams, and separate the syllabus into different units, all based on the class lecture recordings. Moreover, we are hoping to implement a feature which lets users choose the language the quiz will be taken in and the AI converts the language of the lecture to the user's preferred language for the quiz.

Built With

Share this project:

Updates