AI-Driven Customer Support with Multi-Agents

Welcome to the official repository for our hackathon project: AI-Driven Customer Support with Multi-Agents. This project is part of the Hack the Future: A Gen AI Sprint Powered by Data, where we leverage AI-powered multi-agent systems to enhance customer service experiences.

🚀 Project Overview

It start from the src - Source folder Our goal is to build an AI-driven customer support system using multiple agents that can:

• Handle customer queries efficiently
• Provide real-time responses using generative AI
• Escalate complex issues to human support
• Learn from interactions to improve future responses

📌 Objectives

• Implement a multi-agent framework for customer support
• Utilize Ollama-based on-prem LLMs for response generation
• Integrate custom tools such as an API, web scraper, and ML models
• Store and retrieve customer interaction data using SQLite DB
• Develop an embedding model for better response matching

🏗️ Tech Stack

• Programming Languages: Python
• Frameworks & Libraries: LangChain, FastAPI, OpenAI API
• Database: SQLite
• Multi-Agent System: Custom-built using Python frameworks
• Deployment: Docker

📂 Repository Structure

├── docs/             # Documentation & research papers
├── src/              # Source code for AI agents
│   ├── models/       # AI models and training scripts
│   ├── data/         # Customer interaction data & processing scripts
│   ├── api/          # Backend API for customer support interactions
├── tests/            # Unit tests & integration tests
├── .github/          # CI/CD workflows
└── README.md         # Project overview & instructions

🚀 Getting Started

Prerequisites

Ensure you have the following installed:

• Python 3.8+
• Git

Contribution

• Venkatesh R
• Yuvanesh KS