Inspiration
All of us in the group either are diagnosed or know someone diagnosed with ADHD, and have experienced the difficulties of focusing on a task at hand for too long. We wanted to create a tool that helps U of A students, especially those with ADHD, complete assignments on time and with far less stress.
What it does
Our application harnesses the power of LLMs to build optimized schedules for students with ADHD, prioritizing assignments that are due sooner, and breaking up the workload with allotted time for breaks and self-time.
How we built it
We utilized the Python Django framework as a more-or-less full stack solution, making use of its ORM to interact with a SQLite database, and its powerful templating system to generate HTML files. We also styled in CSS and wrote a bit of JavaScript for interacting with the DOM in a few places. We called the OpenAI API (specifically the 4o model) inside one of our Django views, gave it user input, and additional prompts, and then formatted its response as part of our solution. We deployed on an AWS ec2 instance.
Challenges we ran into
For some of us, it was our first Hackathon, which added extra difficulty, even for simply getting started. Past that was the learning curve of adjusting to either a new language, new framework, or new API. Riley had to learn html and CSS for the first time to write most of the display. Joseph had to learn how to access an API in order to interact with Open AI. Toby had to learn how to work with forms in Django to update the database, and learned how to use screen sessions during deployment.
Accomplishments that we're proud of
We're proud that by the end of the hackathon we had a finished project. We are proud that we were able to code the Chat GPT API to consistently output text in a form that we liked. And lastly, we are happy that we were able to create an app that can help fellow students.
What's next for Schedule Bot
Next for Schedule Bot is blocking out times for classes and other commitments, like clubs and sports, and having a way to mark assignments as complete. This creates a far more realistic and usable schedule. Then, with this additional functionality, we can possibly weave Schedule Bot into D2L itself, so all assignments and classes are already built in, requiring way less work to create a complete schedule.
Log in or sign up for Devpost to join the conversation.