Inspiration

The inspiration for this project was the UKG nextGenAI challenge. Why help teachers in particular? We cherish our professors here at Broward College and FIU and would love to one day evolve the idea to come back around and help the very people who helped us!

What it does

Essentially it helps a teacher perform their job more efficiently, providing the ability to create assignments and fulfill other various needs.

How we built it

We build this program solely through Python. Luke worked on the front-end of the program through a GUI library called "Tkinter". I worked on the ai chatting aspect (back-end) to utilize an API provided by OpenAI. Training the bot, having it read files for curriculum summaries, and even down to the way it responds is all done through the back-end Python programming.

Challenges we ran into

We ran into two challenges. One was difficulty using our first AI API, being that there isn't enough info out there on how to set it up; leading us to ChatGPT's API solution. The second challenge was not having enough test data or test cases. As we weren't able to provide our program as a practical solution, we did our best to find course outline files and use those, with the help of a couple professors of course.

Accomplishments that we're proud of

Being able to populate an AI's response into textboxes into a "texting" fashion. Having set up and plugged in the AI into our working program, gaining experience to perform this again in the future.

What we learned

We learned that 24 hours is not a lot of time to get very detailed with a project and is very taxing to try to do so anyway.

What's next for Teacher's Assistant

The goal with Teacher's Assistant would be to improve the GUI, AI response generation time, and build practical use with teachers in the future! We'd like to be able to have the program learn info about a teacher and what they teach in order to tailor information to them when they interact with our program for assistance.

Built With

Share this project:

Updates