NoteMyVideo

Home Page
How it works
Features
Features
Main Dashboard
My Files folder
Chat Room with PDF viewer

Inspiration

While watching a long YouTube tutorial on Model Context Protocol (MCP), I got so focused on understanding the explanation that I forgot to take notes. When the video ended, I realized I’d have to watch the entire hour-long video again just to write notes — which felt frustrating and unproductive.

That’s when the idea for NoteMyVideo came to life: “What if we could automatically generate smart, structured notes from any YouTube video — instantly?”

With this in mind, I wanted to create a tool that helps students focus on learning, while AI handles the note-taking.

What it does

NoteMyVideo is a web app that turns YouTube videos or topic keywords into study materials — in just a few seconds.

Key Features:

YouTube to Notes: Generate both short and long notes from any YouTube video.
Topic-Based Notes: Enter a topic instead of a video and still get AI-generated notes.
Quiz Generation: Get quiz questions to test your understanding of the content.
My Files Section: Access all your previously generated PDFs in one place.
Study Together: Create shared chat rooms where students can study a PDF collaboratively.
Ask Any AI: Tag and chat with different AI models like @llama, @gpt, @gemini, @claude, or @sarvam, and ask questions about the notes directly inside the chat.

NoteMyVideo makes self-learning faster, smarter, and collaborative.

How we built it

We used a combination of frontend, backend, AI, and real-time technologies to bring NoteMyVideo to life:

Bolt AI for building the frontend UI and backend logic efficiently.
Groq Cloud API using LLaMA 3.3 70B and other models for LLM responses.
Supabase for user authentication (Google sign-in) and storing user-generated files securely.
PDF.js for rendering the notes and quizzes in viewable format.
WebRTC DataChannels and Socket.io for real-time collaborative chat rooms.
Custom prompt routing system to switch between AI models based on tags (@model).

Challenges we ran into

Real-time communication was new to us — setting up WebRTC with signaling and proxy routing took time and multiple iterations.
Implementing LLM model tagging (like @gpt) and routing to different APIs required custom backend logic.
Syncing the PDF view across multiple users in real-time without glitches was challenging.
Balancing performance while generating large notes and storing files dynamically required optimization.

Accomplishments that we're proud of

Successfully built a full-stack AI-powered web app from idea to execution.
Enabled real-time collaboration between students with shared study sessions.
Integrated multiple LLMs and allowed dynamic interaction using simple tagging.
Created a clean, user-friendly interface that feels smooth and modern.
Learned and applied WebRTC, Groq APIs, Supabase, and multi-model LLM logic — all in one project.

What we learned

How to use Supabase for user authentication and file storage.
How to integrate and stream responses from multiple LLMs using Groq API.
How WebRTC and Socket.io work to create live, peer-to-peer interactions.
Prompt engineering techniques for note generation, mind maps, mnemonics, and quiz creation.
How to build a real product that solves a real problem for students.

What's next for NoteMyVideo

Add video summarization timeline view to let users jump to key moments.
Fine-tune LLMs for more accurate quiz difficulty levels (easy, medium, hard).
Launch a premium plan with higher usage limits, priority queue, and PDF export customization.
Build a mobile app version so students can use NoteMyVideo on the go.