Skip to content

Latest commit

 

History

19 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 

Repository files navigation

LangGraph Customer Support Automation

Project Overview

An intelligent customer support automation system built with LangGraph that processes incoming emails and provides automated responses using AI agents. The system integrates Gmail API, Google Cloud Pub/Sub, and AWS services to create a complete end-to-end solution.

Business Use Case

Customer support operations often become a significant overhead for businesses, requiring manual intervention for routine inquiries. This project addresses that challenge by:

  • Automating Email Processing: Automatically receives and processes customer emails through Gmail API integration
  • Intelligent Response Generation: Uses AI agents to analyze customer problems and generate appropriate responses
  • Knowledge Base Integration: Leverages company policies and documentation through RAG (Retrieval-Augmented Generation)
  • Order Tracking: Provides real-time order status updates from integrated systems
  • Scalable Architecture: Built on AWS infrastructure for enterprise-scale deployment

Architecture

langgraph-agent-customer-support-automation

Core Components

LangGraph Workflow (langgraph-workflow/)

  • FastAPI Server: Handles Gmail webhook events at /gmail-webhook endpoint
  • Agent Workflow: Multi-step LangGraph workflow for email processing
  • State Management: Tracks email processing state through the workflow
  • Tool Integration: Connects to MCP server for external tool access

MCP Server (mcp-server/)

Dockerized Model Context Protocol server deployed on AWS ECS Fargate providing:

  • Knowledge Base Tool: RAG solution using Pinecone for company policy search
  • Tracking Tool: DynamoDB integration for order tracking information

AWS Infrastructure (aws-cdk-infrastructure/)

  • ECS Fargate: Hosts the containerized MCP server
  • DynamoDB: Stores order tracking data
  • ECR: Container registry for MCP server images
  • VPC: Secure networking setup

About

LangGraph-powered customer support automation with multi-component agentic workflow, AWS CDK infrastructure and MCP server, RAG with Pinecone. AI agents parse emails, fetch tracking data, apply policies, update tickets and generate responses automatically.

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages