LearnInfinity

LearnInfinity — The AI tutor that makes learning infinite, smart, and fun. One platform, every subject, any grade.

Comment 1

LearnInfinity

Inspiration

My Dad has always said that gaps in the foundation of your learning are very hard to fill. This is caused by the constant rate at which schools teach and the fact that the student-to-teacher ratios are much too high for any student to receive personalized teaching. We envision that our product will help all of our users not only learn what they came to learn, but also maybe learn how to learn. This is a very important life skill because you are always learning, no matter how old you are.

What it does

Our product starts by asking our users for their Age, current Grade, and Location(a broad approach is fine, but it is recommended to help with personalization). This is sent to the Api, which now generates a couple of questions that can help the Api determine if the user excels or not at this subject. After the user answers the questions, the Api determines whether or not the user was right, and if they were wrong, then it gives them some links to learn more about this topic. The user is then free to ask the Api any questions that they have to help fill their knowledge gap.

How we built it

We built it with one goal in mind: to boost. whether it is to boost someone's learning, their grasp of a topic, or maybe just their curiosity. With this, we moved forward and started making the backend. The backend called the Api with its prompt engineering and system instructions. which helped the Api create the best response possible for the user. Once we finished testing it with Postman, we made the frontend which made Axios JSON requests to the backend which wrapped it with prompt engineering and system instructions and sent it to the Api. When the Api responded the backend would make it into a perfect JSON message and send it back. which would then finally be displayed to the user.

Challenges we ran into

One of the challenges that we ran into was making it personalized at a high level. This was very difficult because a lot of system instructions and prompt engineering were involved. Another problem we faced was time management, this is because we had a couple of other activities planned during this time, and it was difficult to give it all the time it really needed.

Accomplishments that we're proud of

We are very proud of how highly we took our level of personalization. From the start, this was a very big aspect of our project. We also think that our use of AI in a way that helps and does not deter learning is also a very big accomplishment. Lastly, I think that after the limited time we had I think that we have done a wonderful job in the CSS, and the frontend looks beautiful.

What we learned

We have learned a great deal during this hackathon including how to make backend servers. When we first started the very first thing we did was make a frontend server that would call the API(there were no actual pages just yet). We faced a couple of CORS issues and quickly pivoted to the backend idea. We learned that backend servers can be made on Python Django, so that's what we went with. We learned that you can quickly send requests from frontend to backend, backend to API, and back very quicky with AXIOS.

What's next for LearnInfinity

I believe that the next step for LearnInfinity is to either make or train its very own model. This is because I believe that since we are using Gemini there are possibilities of someone using it in ways that are not helpful to their learning. A couple more steps that we plan on taking are, to expand the curriculum of the AI, add ways such as text-to-speech that can help disabled people, and lastly roll out a alpha model for students and teachers to use.

Built With

Share this project:

Updates