Lear-Your-Way

Inspiration

I noticed that sometimes students don't struggle with the concept itself — they struggle with how it is explained. A textbook may not make sense, while the same idea can click through a simple explanation, analogy, visual, or step-by-step breakdown.

That inspired me to build Learn Your Way: an app where the learner can change the explanation instead of struggling with the same format.

What it does

Learn Your Way lets learners paste or upload learning material and get AI-powered explanations in four ways:

Simple
Analogy
Visual
Step-by-step

If an explanation doesn't click, they can use "Explain it differently" to try another approach. It also supports PDF and image uploads, practice questions, quizzes, and a rethink flow for concepts they missed.

How we built it

I built the frontend using HTML, CSS, and JavaScript, with a Python/Flask backend deployed on Vercel.

I used NVIDIA's AI API for text and vision capabilities, pypdf for PDF text extraction, Pillow for image processing, and the browser's Web Speech API for the Listen feature.

Challenges we ran into

One of the main challenges was making the AI integration reliable in deployment. Image processing also took longer than normal text requests, which caused timeouts during testing.

I also had to make sure API keys stayed on the server and were not exposed in the frontend or GitHub repository.

Accomplishments that we're proud of

I'm proud that the project became more than just an AI chatbot. The "Explain it differently" flow gives the learner control over how they approach a concept.

I also got the full flow working from learning material → explanation → practice → quiz → rethink.

What we learned

I learned that building an AI product isn't just about connecting an API. The way the AI is presented and how the learner interacts with it matters just as much.

I also learned about API integration, server-side environment variables, PDF and image processing, vision models, deployment, and building a complete learning workflow.