Inspiration
We noticed that AI is becoming a large part of education. In particular, while sitting in math lecture, my friends and I started using LLMs whenever the professor made a statement we didn't understand. We would ask the LLM to re-explain the point right away. We then realized there was no streamlined process to quickly ask and store questions you have while learning. Thus, we created NeoLearn.
What it does
NeoLearn gives everyone easy access to an LLM prompted to maximize their learning. By structuring courses with key points, students can store the questions they had for every topic. We also have a chat to ask questions about problem sets, where the LLM will guide the student through the question, rather than give the answer immediately. We also have the option to autogenerate multiple choice quizzes.
How we built it
We used NextJS and TailWindCSS for the backend and google cloud firebase as the database. For the LLM, we used the Google Gemini API.
Challenges we ran into
We spent a lot of time working on the rendering of the LLM output to make it nice for a student to study.
Accomplishments that we're proud of
We're proud we made an app that provides useful insight into education and created a product we ourselves will used going forward.
What we learned
We learned how to use clever prompting and clean software to make certain use-cases of LLMs more tractable and easy to use.
What's next for NeoLearn
We want to add an easy way for teachers to create custom lessons for their students through out platform.
Log in or sign up for Devpost to join the conversation.