TutorGPT

adding notes
generated question

Inspiration ⭐

As the summer approaches, we find ourselves unable to concentrate on studying. We wanted to create a tool that would allows students to enjoy studying!

What it does 🚀

TutorGPT is a review game with AI-powered question generation that makes studying fun!

How we built it 🔨

We built TutorGPT using Python, JavaScript, HTML, and CSS. Team members collaborated over Github. We followed a three step process of ideation, implementation, and iteration.

Ideation
We list out a series of ideas, then merge and modify existing ideas until we create a niche product that serves a very practical purpose--TutorGPT. We hope that you find TutorGPT effective in your study life, and we are actively working to improve the app to become even greater in the future.
Implementation
We grinded for two days, called until 1:08 AM (the GitHub history shows), and almost failed our chemistry exams (we would have if not for TutorGPT!) to complete the app, but what was done is done. We are proud of our product, the learning experiences, and the connections we created during this event.
Iteration
We reviewed all the amazing work we completed throughout this event, and we asked ourselves: what can we do better? Building upon our initial project, we upgraded TutorGPT to analyze images, further improving the study experience.

Challenges we ran into 🤔

We decided to use XMLHttpRequests instead of forms, which Django usually uses, to transmit data from the frontend to the backend so as to avoid reloading the page. Unfortunately, this approach was more poorly documented so we ran into multiple errors. Curating the correct prompt for ChatGPT integration was also a challenging process, but we strategically attempted various phrases and devised the prompt in the finished product. We will seek to improve TutorGPT by optimizing the prompt more in the future.

Accomplishments that we're proud of 😺

We are proud of creating a program that successfully works with ChatGPT. One feature we are most proud of is that TutorGPT can accept images and generate questions from them.

What we learned 🧠

Throughout this project, we learned how to use the OpenAI API to generate useful practice questions to solidify a student's understanding of a topic. In addition, we learned how to integrate this powerful API with Django, a high-level Python web framework. This allowed for a fast and easy interface that automatically stores the generated questions and answers.

What's next for TutorGPT 🏅

We hope to continue improving TutorGPT in the future! Our goal is to receive audio inputs, such as lectures, and generate questions from them. Another goal is to have TutorGPT offer explanations for answers.