
Inspiration
I have always had a lot of trouble getting into new topics of learning. There is a huge lack of resources and structured learning pathways to get into new material. This is exceptionally hard for getting free resources in a bundle to learn from, especially when you are unfamiliar with the technical jargon terms. So I decided to make something that is able to help spread awe in
What it does
TreeEd is an AI powered system. Think of it as your personal resource collector! With just a A LITERAL single click, it runs AI processes to bring personalized course structures to you. As it is a structured learning pathway, it brings the benefit of both self-learning and courses.
How we built it
TreeEd utilizes a naive approach with a powerful LLM model via inferencing to fetch important topics in the listed order of course structure. These topics are assigned a difficulty level and on the basis of that, and are arranged in a system a gradual increase in difficulty level of learning. Following this, using a sophisticated webscraping technology, it scrapes top courses/resources from the web on how to learn each specific topic. This is a POC AI system which can be further scaled up into a full scale AI model. Even further, it allows filters, such as looking specifically for youtube videos. You can also enlist further explanation by an LLM to help you delve deep into the topic you are learning.
Challenges we ran into
A really important problem that we ran into was time constraints. However, we were able to persevere through it and finish the project in the nick of the time.
Accomplishments that we're proud of
I am extremely proud of the project that I have created, and the idea it stands for. It has the potential to help ignite the curiosity for thousands of kids all across the globe and that is something really exciting.
What we learned
I was able to look into automated prompt engineering and sophisticated webscraping technologies. I was also fascinated by how this affected learning patterns and would love to learn more on that.
What's next for TreeEd
This is just the beginning for TreeEd. This was a naive approach to this problem, and I believe there is great potential for creating sophisticated AI systems using vector space of words, or even training AI model to create and alter personalized resource packs. This could be helpful for self-learners, or for educational institutions.
Built With
- ai
- python
- streamlit



Log in or sign up for Devpost to join the conversation.