Skip to content
This repository was archived by the owner on Jan 31, 2026. It is now read-only.

chandlerNick/langchain-multi-agent-demo

Repository files navigation

Agentic AI Email Automation System

Built with LangGraph

This project implements an agentic AI workflow to automate email handling and responses using LangGraph.
The system intelligently reads incoming emails, classifies their intent, drafts responses, and sends them - with optional human-in-loop review for critical messages.

It utilizes GPT-OSS-120B as the underlying LLM which we deploy on the BHT compute cluster's Nvidia A100 80GB GPUs.

Further details

For a quick overview of the project (including photos of UI), look at Agentic_Email_Automation_Presentation.pdf

If you want to replicate the workflow (and have sufficient resources) examine the directories src/multi-agent-system and src/streamlit_ui as well as the .env.example and Makefile.

Note that the system primarily follows this tutorial with some modifications and plumbing / ops done to serve the LLM, create a front end, and wrap the agent in a service.

About

We utilize LangChain detailing how it works and give an example multi-agent application.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors