LangGraph: Hierarchical Agents - How to build Boss & Subordinate Agents

In this video I will show you how to build hierarchical agents with LangGraph. Subordinate Agents will ALWAYS Report to a Boss Agent, who than makes a final decision on a specific topic. We will create a fictive news agency.

Timestamps:
0:00 Introduction
0:45 LLM Setup
5:51 LangGraph - Nodes & Edges
12:31 Result

#langchain #langgraph #agents

[iwswordpress]

This is excellent! Have got your Udemy course...

[sempercrescere]

Amazing video!
I have one quick question that I hope you could help with. Some of your node functions require global variables (i.e. the llm) that are not passed as an argument. Some people might argue that the use of global variables isn't recommended.
I was wondering if there is a way that these node function s take more than the state as an input, i.e. both the state and some other inputs (e.g. the LLM)?

[pabloguatibonza]

Great video!
I would like to see a video about a customer support chatbot(with Q&A and RAG)

[Prathameshhh_]

Great sir, Thank you for the video.
In the video you have used llm for multiple tasks defined multiple times like postability, translator and so on.
is this approach valid and acceptable in production grade agentic systems?
i

[DanielWeikert]

Have you seen (or use it yourself) that LangGraph is used in company environment in production? Thx

[oluwaseunakinropo]

i really want to see a langgraph fastapi version from you ser

[miguelangelmurillo]

Why has gpt-4o changed the way of using calling function, which is essential for an agent? How to integrate into these systems