Inspiration
Studying for exams ourselves, we understand the value of study resources like past exams. We were inspired by other apps like Turbo AI, and we wanted to expand on a problem we saw with their products. They generate random questions for learning, when a better, yet more expensive option would be to select questions specifically tailored to a course, improving user experience and studying effectiveness.
What it does do
Our desktop app takes in a users course syllabus as input, and using Google's Gemini, we extract a generalized course name as well as covered content. Then, with these keywords, we used SerpAPI to scrape for past exams from top universities offering the same course. Then, we compiled these questions into a database, and allowed the user to create customized parameters for their exams, creating unique and repeatable practice material.
How we built it
We built the desktop app frontend using Electron for the application and Tailwind for UI design. For our locally hosted servers we used FastAPI and Python to implement search, scraping, and classification pipelines. Finally, for classification, we used Google's Gemini 2.0 Flash, ensuring a balance between response time and token usage. For storing data we used SQLite, handled with Python and Typescript.
Challenges we ran into
Although we had a clear vision of what we wanted to build throughout the hackathon, we ran into problems with connecting between the frontend and backend, with it being our first time using Electron and FastAPI. Through experimenting, research, and -> Claude AI <-, we were able to successfully hook everything up.
Accomplishments that we're proud of
We're proud to present a product that could be helpful to many people, while being able to enjoy the design and building process.
What we learned
With this being our first time working under such a tight time restriction, we improved our skills in time management and collaboration, ensuring that team members kept up with tasks and maintained good communication. Additionally, this experience strengthened our app-building skills, along with user-oriented design.
What's next for LearnIO
With LearnIO we see a lot of potential for new features and optimizations, as well as a possible profit model due to it's scalability. We will be able to move it from a desktop application to an online webpage, making it more accessible and user-friendly.
Log in or sign up for Devpost to join the conversation.