Inspiration

We wanted to build this so that people all over the world can learn. There should be no reason to require something like electricity or WiFi to learn. At the very least, everyone should have access to elementary level academics. This was the problem we wanted to tackle for this hackathon.

What it does

This application has set out the necessary items to learn for elementary level mathematics and english. Not only that, it has an offline AI built into it so that it can make your learning experience more well-rounded.

How we built it

We built this tool using Django as the back end and the front end with typescript. We also implemented OpenAI as the chatbot since Ollama took too long to respond and would inevitably time out.

Challenges we ran into

The main challenge we had was incorporating AI into the chatbot. At first, we thought about using PyTorch and Transformers, but a laptop would take too long to process those Large Language Models. In the end, we decided to use OpenAI for the time being, and figure out a different LLM that is lighter on processors to substitute as the offline AI.

Accomplishments that we're proud of

We are proud of how well this project came together. From conceptualization, we had an idea of what to do built the fact that it was able to come together is something we, as a group, are extremely proud of.

What we learned

What's next for Lumi: An Offline Studying AI

Built With

Share this project:

Updates