https://github.com/djdurga/langgraph_demo

https://github.com/djdurga/langgraph_demo

Last synced: about 1 year ago
JSON representation

• Host: GitHub
• URL: https://github.com/djdurga/langgraph_demo
• Owner: Djdurga
• License: mit
• Created: 2025-06-29T09:09:23.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2025-06-29T09:14:52.000Z (about 1 year ago)
• Last Synced: 2025-06-29T10:21:38.598Z (about 1 year ago)
• Language: Jupyter Notebook
• Size: 56.6 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# 🔁 Exploring LangGraph: Agentic Workflows with Python

Welcome to my AI experiments with **LangGraph** — a Python framework for building **stateful, multi-step, and agent-like applications** powered by large language models (LLMs). This repository captures my journey in understanding and applying LangGraph to create structured, intelligent, and flexible AI workflows.

---

## 🌐 What is LangGraph?

[LangGraph](https://github.com/langchain-ai/langgraph) is an open-source Python library that lets you build **dynamic computation graphs** where each node can represent:

- a language model call,

- a tool invocation,

- a data transformation,

- or an agent decision.

It is inspired by **state machines** and **dataflow programming**, offering native support for memory, conditional logic, and complex multi-step reasoning — something traditional LLM wrappers often lack.

---

## 🧠 Why I'm Exploring LangGraph

I’m diving into LangGraph to:

- Move beyond single-step prompts and unlock **structured reasoning** with LLMs

- Build **modular pipelines** for agentic tasks like tool use, retrieval, and memory

- Understand how AI agents can maintain state and make informed decisions

- Experiment with **agent workflows**, chatbots, and RAG (retrieval-augmented generation)

LangGraph provides a fresh way to think about AI development — focusing on flows, not just prompts.

---

## ✨ Key Features of LangGraph

- 🧩 **Composable Nodes**: Each step in your graph is a modular function or model call

- 🔁 **Stateful Workflows**: Easily pass memory and data between steps

- 🧠 **LLM-Aware Agents**: Design agents that can reason, plan, and interact

- 🔀 **Conditional Branching**: Control logic flow with dynamic decision-making

- 🔄 **Support for Loops & Multi-Agent Patterns**

---

## 📚 Resources to Learn LangGraph

- 🧭 [LangGraph Documentation](https://docs.langgraph.dev/)

- 🛠️ [LangGraph GitHub Repo](https://github.com/langchain-ai/langgraph)

- 🎓 [LangChain Tutorials on YouTube](https://www.youtube.com/@LangChain)

---

## 🛠️ Next Steps

This repository will include:

- [ ] Examples of simple LangGraph workflows

- [ ] A mini-project using LangGraph and LangChain together

- [ ] Learnings, notes, and gotchas from real use cases

Stay tuned!

---