Project Info

StudyHubby

This project did not submit a demo video on Devpost.

Inspiration

: Oftentimes, we find studying for course content hard to follow and wish we always had some extra assistance when it came to studying. Lot's of AI tools existed but they were all over the place and we just wanted to make the one stop shop for studying. And that is why we made Study Hubby.

What it does

: It takes in a transcript from your video lectures and then allow the user to have conversations with the chatbot to help students learn much better. It provides options to the user such as a helpful lecture chatbot, summarize key points in the lecture, and also provides a quiz system to further the users understanding of the topic. We also implemented a test system where the user has to explain the complex topic to a 5 year old in order to really gauge their understanding.

How we built it

: We created our backend using a mix of fastapi and Python and also utilized the Gemini API service in order to create the AI generated content that will accelerate the student's learning. We created the frontend in NextJS and Tailwind CSS.

Challenges we ran into

: Building the backend was quite difficult because we had to use fastapi and we never used it before and the learning curve was quite steep. Another challenge was the fact that we couldn't implement all the features that we wanted due to time constraints. We also learned that to use the caching function of Gemini, we needed a longer text transcript which was a challenge because not all course content will be long enough to fill it up so we decided not to use caching.

Accomplishments we're proud of

: The fact that the basic AI functions were able to work and even the small parts of our code like the quiz parser, we are proud of because we put a lot of effort into it.

What we learned

: We learned a bit about fastapi and how to implement it into our own apps to have a python based backend solution. We also learned the limits of AI generated content and how to make it more relevant to education.

What's next

: We have a lot in the tank. For starters, we want it to be the center of all learning the user will engage in as we want to add multi user support and multi course support so the user does not need keep adding their lecture transcripts every time.

Analysis

Compare with all teams

View

Metric

KYUTI3
18
Buyantogtokh Khurelbaatar
2

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

CSSIn code
Next.jsIn code
ReactIn code
Tailwind CSSIn code
TypeScriptIn code
FastAPIClaimed
Google GeminiClaimed
PythonClaimed

5 of 8 appear in the indexed code. 3 claimed on Devpost could not be matched to code, which may simply mean the tool leaves no trace in the repository.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

224 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.