Generative AI with LangChain - Second Edition

3

Building Workflows with LangGraph

So far, we’ve learned about LLMs, LangChain as a framework, and how to use LLMs with LangChain in a vanilla mode (just asking to generate a text output based on a prompt). In this chapter, we’ll start with a quick introduction to LangGraph as a framework and how to develop more complex workflows with LangChain and LangGraph by chaining together multiple steps. As an example, we’ll discuss parsing LLM outputs and look into error handling patterns with LangChain and LangGraph. Then, we’ll continue with more advanced ways to develop prompts and explore what building blocks LangChain offers for few-shot prompting and other techniques.

We’re also going to cover working with multimodal inputs, utilizing the long ...