AutoGen Multi-agent Conversation and Recall using Memori

This repository contains a Jupyter Notebook demonstrating how to create memory-enhanced AutoGen AI agents using the Memori memory engine. The example builds a consulting team of AI agents that can remember and build upon past conversations, making interactions more intelligent and context-aware.

Features

• Persistent Memory: Agents remember conversations across sessions.
• Multi-agent Collaboration: AI agents work together to solve client problems.
• Real-world Scenario: Simulates a software consulting team assisting a client.

Getting Started

1. Install the required dependencies:

   pip install memorisdk autogen-agentchat "autogen-ext[openai]" python-dotenv

2. Set up your OpenAI API key:

   • As an environment variable: export OPENAI_API_KEY="sk-your-key-here"
   • Or in a .env file: OPENAI_API_KEY=sk-your-key-here

3. Open the notebook agentchat_groupchat_memory_using_memori.ipynb and follow the steps to:

   • Initialize the memory system.
   • Create AI agents with distinct roles.
   • Simulate client conversations.

Use Cases

• Customer Support
• Project Management
• Educational Tutoring
• Medical Consultation

Learn More

Explore the notebook to see how Memori enables smarter, memory-enhanced AI interactions.