gpTA: AI Canvas analyzer

gpTA: AI Teaching Assistant Chatbot logo

Inspiration

The inspiration behind gpTA was to streamline the process of accessing and interacting with course information by leveraging the Canvas API, LangChain API, and GPT-4 model to provide students with a comprehensive AI-driven educational experience.

What it does

gpTA collects course data from the Canvas API, processes it through the LangChain API for storage and analysis, and then allows users to communicate with a GPT-4 chatbot through a user-friendly React application, enabling personalized educational support.

How we built it

gpTA was built through a multi-step process involving Canvas API integration, LangChain API utilization for data storage, and the development of a React application for user interaction. The core of the system involves the integration of these components.

Challenges we ran into

Some challenges we ran into included fetching data from Canvas's API, figuring out how to connect the virtual files returned by Canvas's API into LangChain, and acquiring a GPT-4 API key to use LangChain. In addition to this, half of our teammates were not able to come into the building until 6:00 PM on Saturday, which took up much of our development time.

Accomplishments that we're proud of

We are incredibly proud of learning new technologies we have never worked with before, successfully integrating multiple APIs, developing a functional data pipeline, and building a streamlined and aesthetically pleasing frontend. By the end, we were able to successfully create an accessible and efficient educational tool that we believe will be incredibly helpful for every student who uses Canvas for their classes.

What we learned

Through the development of gpTA, we gained valuable experience in API integration with the Canvas and LangChain APIs, data processing, and creating a user-friendly interface. This project may have deepened our understanding of AI-powered educational tools.

What's next for gpTA

The future of gpTA could involve storing user conversations on the cloud to further streamline their learning process, further refining the AI capabilities of the GPT-4 model, and expanding its compatibility with other modes of information, like lecture videos, as well as other educational platforms to offer a more comprehensive and personalized educational support system.