🚀 Course Maker AI

🎯 Inspiration

Learning is everywhere, but structure is rare.

We realized that while there’s an explosion of great content on YouTube, Twitter, Reddit, and blogs, most of it is scattered, unstructured, and hard to retain. We wanted to build a tool that could take this rich, unstructured content and turn it into structured, AI-generated courses — fast.

Imagine turning a 10-minute video or a long blog post into a complete mini-course with lessons, summaries, and even AI-generated videos. That’s what sparked Course Maker AI — a way to democratize course creation and accelerate how we learn from the internet.

💡 What it does

Course Maker AI transforms raw content into structured learning experiences using AI.

🔗 Accepts input links: YouTube, PDFs, blogs, Twitter threads, or Reddit posts.
🤖 Extracts relevant content (e.g., YouTube transcripts or blog text).
🧠 Sends the content to a powerful AI model (via Groq API) to generate:
- A course title and description
- Multiple structured lessons with summaries and bullet points
- (Coming soon) AI-generated videos for each lesson using Tavus
📦 Outputs a ready-to-publish course preview
🧵 Supports forking, sharing, and personalization of learning paths

🏗️ How we built it

Frontend: Built with React + Tailwind CSS, styled for smooth animations and responsive UX
Backend: Node.js + TypeScript server to handle content extraction and Groq-based course generation
AI Integration: Used Groq’s fast inference API (Mixtral-8x7b model) to generate structured content from plain text
Extraction Layer: Custom content parsers for YouTube, PDFs, blogs, Reddit, and Twitter — with support for mocking in dev
Dev Tools: Vite, tsx, and concurrently to run full-stack locally with hot-reload

🧗 Challenges we ran into

🧪 Reliable content extraction: Handling inconsistent formats from YouTube, blogs, or social media required custom extraction logic.
🔐 Groq SDK availability: The official SDK wasn’t available on NPM, so we had to mock or hit the raw API ourselves.
⚙️ Synchronizing backend and frontend: Ensuring seamless end-to-end flow between user input and course output.
🧵 Maintaining flexibility: Supporting five input types while keeping the UX simple and scalable.

🏆 Accomplishments that we're proud of

🧠 Built a working system that converts raw internet content into structured learning.
⚡ Integrated Groq’s blazing-fast model for near real-time course generation.
🎨 Designed a polished UI/UX that makes course creation feel magical.
🔄 Enabled modularity and extensibility with support for future integrations like Tavus (for AI-generated video) and RevenueCat (for monetization).

📚 What we learned

✨ The power of fast inference models (like Mixtral on Groq) in educational tools.
🧩 How to architect AI workflows with clean separation of concerns — extraction, generation, presentation.
🛠️ Practical skills in React, Node.js, TypeScript, Tailwind, and working with LLMs.
🤝 How to collaborate effectively as a 4-person team in a time-boxed hackathon.

🔮 What's next for Course Maker AI

🎬 AI video generation using Tavus for each lesson
🌍 Publish and share courses with custom URLs and social embeds
🧑‍🤝‍🧑 Collaborative learning paths: Let users remix and fork courses together
💸 Premium features via RevenueCat + Algorand NFT-based ownership tracking
🤖 Voice cloning + multilingual courses powered by ElevenLabs

Course Maker AI is just the beginning. We're reimagining how the world learns — one link at a time.