LangChain vs LangGraph: which to use for Agents?

My AI courses: https://edwarddonner.com/curriculum In the courses: AI Engineer Core Track includes LangChain AI Engineer Agentic Track includes LangGraph Here's the Google Colab with the walk-through: https://colab.research.google.com/dri... 00:00 The puzzle 00:35 The 4 layers explained 03:51 Hands-on in Google Colab 05:34 LangChain Core in action 09:13 LangGraph in action 11:22 The agent framework in action 12:11 The surprising reveal! 12:57 And now.. Deep Agents in action 15:08 The mission for you I hope this finally solves the head-scratcher of LangChain vs LangGraph vs create_agent vs Deep Agents. The perplexing number of ways to create an Agent in the LangChain ecosystem. Let me know if you have any more AI puzzles for me to solve!