Have a 5-hour YouTube video but no time to watch it? This LLM application pulls the main topics and explains to you like you are 5, so you can catch up in just minutes.
Design Doc: docs/design.md, Flow Source Code: flow.py
Try running the code in your browser using the demo notebook.
-
Built With Pocket Flow, a 100-line LLM framework that lets LLM Agents (e.g., Cursor AI) build Apps for you
-
Check out the Step-by-Step YouTube Tutorial:
 NVIDIA CEO Jensen Huang's Vision for the Future |
 DeepSeek, China, OpenAI, NVIDIA, xAI, TSMC, Stargate, and AI Megaclusters |
 Demis Hassabis – Scaling, Superhuman AIs, AlphaZero atop LLMs, AlphaFold |
 Elon Musk: War, AI, Aliens, Politics, Physics, Video Games, and Humanity |
|---|---|---|---|
 In conversation with Elon Musk: Twitter's bot problem, SpaceX's grand plan, Tesla stories & more |
 In conversation with President Trump |
 Jeff Dean & Noam Shazeer – 25 years at Google: from PageRank to AGI |
 In conversation with Tucker Carlson, plus OpenAI chaos explained |
 Jonathan Ross, Founder & CEO @ Groq: NVIDIA vs Groq - The Future of Training vs Inference |
 Volodymyr Zelenskyy: Ukraine, War, Peace, Putin, Trump, NATO, and Freedom |
 Sarah C. M. Paine - Why Dictators Keep Making the Same Fatal Mistake |
 Satya Nadella – Microsoft’s AGI Plan & Quantum Breakthrough |
 Full interview: "Godfather of artificial intelligence" talks impact and potential of AI |
 The Stablecoin Future, Milei's Memecoin, DOGE for the DoD, Grok 3, Why Stripe Stays Private |
 The Future Mark Zuckerberg Is Trying To Build |
 Tucker Carlson: Putin, Navalny, Trump, CIA, NSA, War, Politics & Freedom |
-
Implement
call_llminutils/call_llm.pyso it takes a string and returns a string. -
Install the dependencies and run the program:
pip install -r requirements.txt
python main.py --url "https://www.youtube.com/watch?v=example"- When it’s done, open output.html (created in the project folder) to see the results.