Inspiration
As CS students, we know how much of a hassle it is to get someone to advise us on which classes to take, and how sometimes the advice of these advisors usually ends up with us having hectic semesters with the course or very light semesters where you feel like you wasted your time.
What it does
This web-based program allows users to input their information such as their course load and what they want out of the schedule and our algorithm will return a schedule showing all the different courses they need to take
How we built it
We built a front end using HTML and CSS, using Flask to integrate the front end to the back end. For the back end, we built it using a combination of a Chatgpt API as well as some Python code to help with integration. We built a database of classes, which allowed us to check for prerequisites as well as credit value. We used computer science as our example major and built the database based on the flowchart of recommended courses. We used chatGPT to analyze the data as well as take input from the user to allow for personal preferences such as wanting to take a class in a certain semester or using AP credit to transfer out of a class.
Challenges we ran into
3 major challenges we ran during this hackathon. The first was trying to implement GPT4 into our program. We originally wanted to feed the flowchart of every major and it would construct a database to hold the info, thus making it easier to scale, but since this was the first time we were building something with an API, we didn't want to risk 40 dollars on the cost of the GPT4 API just to ultimately fail in doing with the implementation. We got the chatGPT 3.5 API to work but then struggled with the integration into the front end but we eventually succeeded. The second was deploying to AWS. We initially wanted to use AWS to host the database but had trouble when implementing it as there were some syntax issues when converting the database. The main challenge we ran into was our complete lack of experience in all aspects of hacking. This is our first time attending a hackathon so we spent a lot of time just researching how to implement all of our ideas while taming our ideas down to a product we could realistically build.
Accomplishments that we're proud of
We managed to learn and implement all the skills used in this product such as API, flask, and integration of chatGPT. This was no easy feat as all of us involved are freshmen with minimal experience. We also managed to make a working product within the time limit.
What we learned
We learned a lot in these 24 hours such as API, how to implement AI into the backend, and Flask.
What's next for Lion's Advisor
We want to potentially make the backend a little cleaner so it would help with scaling, we could potentially collaborate with Penn State to integrate this into Lionpath I feel this would be a great asset to the university as well as students.
Log in or sign up for Devpost to join the conversation.