Agentic AI Intro using LangGraph

This is a very basic introduction to Agentic AI, aimed at helping you understand the foundational concepts and practical tools used to build agents that can reason and act on our behalf.

We are currently experimenting with a agent workflow tools: LangChain (specifically its graph-based framework, LangGraph). However, the concepts here are not limited to LangChain — there are other frameworks out there, such as:

• CrewAI
• AutoGen
• Qwen-Agent
• OpenAI Agents SDK

This is not intended to be a complete or exhaustive guide. There is a lot happening in the Agentic AI space, with many powerful tools and evolving best practices. What we present here is just a starting point. There’s much more to explore beyond this intro — including topics like memory management, human-in-the-loop, and Model Context Protocol (MCPs)

Learning Flow

1. Agentic AI Basics

We begin with the core concepts of Agentic AI, using a foundational source from Anthropic that explains the structure and goals of agents — how they plan, reason, take actions, and interact with tools.

Article source

2. LangGraph Essentials

Next, we cover the LangGraph, a framework for building composable, reactive agents using LangChain.

Through basic, hands-on examples, we explore how to:

• Build a state graph
• Handle message flow
• Incorporate tools

To be able to run the examples, you should:

• clone this repository
• add the required keys in the .env file.
• create a virtual environment (e.g python -m venv env)
• activate the virtual environment using the activate script
• install the packages present in the requirements file (e.g. pip install -r requirements.txt)
• opening the notebooks using your preferred IDE (e.g. vscode, pycharm, vi)

3. Exercise – Trello Task Manager Agent

In the exercise, we create a task management agent that can:

• View tasks from a Trello board
• Add new tasks
• Mark tasks as done

We give the agent a set of tools and let it decide what actions to take based on the user's intent.

Happy experimenting!

4. Additional resources:

https://www.cohorte.co/blog/unleashing-the-power-of-langgraph-an-introduction-to-the-future-of-ai-workflows https://www.ionio.ai/blog/a-comprehensive-guide-about-langgraph-code-included https://anderfernandez.com/en/blog/agent-systems-with-langgraph/