Bullseye

Running out of material to study for your final? Bullseye's got you covered. Put in your practice tests and get new questions that match the structure and difficulty of your material.

Comment

Inspiration

Going through high school, we had some ridiculously difficult teachers that gave a limited amount of practice material for their tests. On top of that, finding resources online that represented the difficulty of their curriculum was very challenging and time-consuming. We decided that there should be a tool that gives students the practice they need without wasting their time, which brought us the idea of Bullseye.

What it does

Bullseye takes in practice tests and material from students and uses AI to provide new questions that match the difficulty and structure of the test, gives feedback on student answers, and grades the students on their mastery of certain topics.

How we built it

We started the build process with a plan of the whole structure for the project and how it will carry out all of its desired functions. We decided on a web-based project that would be split into three parts: one where the user uploads their study material, another where the user answers the AI generated questions, and a third one to keep track of the user's mastery. Based on that we determined how we will go about creating this project. Once we decided on our build process, we started using the React framework to program the basic functionality of the project, utilizing Node.js for the backend server and Vite for the client server. To parse the contents of the input PDF provided by the user, we used Aryn. Also, we used OpenAI LLM API's to generate questions based on the data provided by Aryn, analyze student answers, and grade the students on mastery. Additionally, we used Tailwind.css to design the user interface of the whole project. We also utilized ChatGPT to aid us in the process of constructing this project.

Challenges we ran into

We had a hard time getting the client and server to interact seamlessly and getting some of the basic website functionality, like the PDF upload prompt and buttons to navigate the different webpages. Some other challenges that we encounter include the integration of Aryn and the Open AI API into the website to do things like parse PDFs and generate content.

Accomplishments that we're proud of

In general, we got a functional backend and frontend for our website, which was really impressive given our lack of experience. We are also very proud of how we implemented a wide range of APIs cohesively into our website to create a useful product, which seemed extremely daunting at first.

What we learned

We gained an immense amount of experience in web development. Some of the things we learned include the implementation of frontend and backend servers, the usage of tools like Aryn and Open AI, and the debugging process for programs.

What's next for Bullseye

We plan on designing this website to handle a large amount of users, through the implementation of a database that will handle all user data. We also plan to clean up the UI and develop a more multi-faceted mastery analysis algorithm.

Built With

Share this project:

Updates