Inspiration
Ever scrolling through dozens of files to find a barely-mentioned information? We understand, and want to help you to save time with Study Buddy - a cute study-with-me AI made by students, for students.
What it does
You can:
- ask questions about concepts/definitions/information in your files
- ask the AI about anything on the Internet, or to solve an exercise in your textbook!
- have quizzes (MCQs/open questions) about any topic in your files
- summarize a topic in your files
- or switch to Asian Mom mode for extra sassiness!
How we built it
We initially envisioned a Discord Bot that takes care of all lecture notes uploaded on the channel, and answer questions based on them. But after getting introduced to Retrieval-Augmented Generation (RAG), our team is excited to embrace the challenge and refocus our resources to the now study-with-me Study Buddy!
Learning from this tutorial link as the spine for our software, we came up with supportive features and experimented with the settings to produce the best accuracy we can during the two days. As a creative twist, we added the Asian Mom mode, part to entertain our users, part to keep the programmer from sleeping past 2am.
Challenges we ran into
We are a team with no AI expert, for some this is their very first exposure to working with AI. Furthermore, for most of us this is our first ever hackathon, so we needed time for research, debating on the best course of action, and most importantly learning the technical skills needed for our vision. However, we are very proud of each other for not only accepting a change in vision mid-way, but being able to pull through them all. A lot of love were poured in Study Buddy, so we hope you enjoy them too!
Accomplishments that we're proud of
Very proud of our team for climbing through the learning curve, as well as the bravery to welcome the new challenge! Go team!
What's next for Study Buddy
In the future, Study Buddy will return with better accuracy, error-handlings, accept more file types - images, power-points, etc. But most importantly, to realign with our initial goal - to bring the program to online group discussions!
Shoutout
A big thank you to the organizers of DevEd 3.0! You understood our struggles and stepped in to support, to that, we are extremly thankful of! Also thank you for all your efforts on this hackathon, thanks to you, many had the opportunity to branch out of their comfort zone. Alongside, gratitudes to Hadi for guiding us on what RAG is and how to utilize them!
For Study Buddy development team, thank you for coming together and persevere until the end through ups and down. All the best on your career afterwards!
Log in or sign up for Devpost to join the conversation.