Learn how to leverage LangGraph's new Functional API to build AI workflow agents with minimal code changes. In this tutorial, we transform an agent implemented in Python into a LangGraph-agent using a few simple decorators. We clearly show what LangGraph enables, including persistence for short-term memory, human-in-the-loop, streaming, tracing / debugging, and long-term memory. Docs: https://langchain-ai.github.io/langgraph/concepts/functional_api/ Video Chapters: 00:00 Introduction to LangGraph Benefits 01:00 Building a Vanilla Python Agent 02:45 Introducing LangGraph's Functional API 07:40 Adding Human-in-the-Loop Capabilities 11:30 Time Travel and State Management 14:00 Implementing Long-Term Memory 18:00 Conclusion and Benefits Review Video Notes (notebook referenced): https://github.com/langchain-ai/langgraph/blob/6ed63ba8fc73fc1e0d205b1353d434c013defae8/docs/docs/tutorials/functional_api/react_functional_api.ipynb