LeadGenAI

LeadGenAI is a workflow to generate and optimize personalized lead generation emails using two Large Language Models (LLMs), orchestrating the content for maximum effectiveness. It leverages LangGraph to model agents and tools.

Features

• Dual-Language Model Workflow: Uses two different OpenAI models (gpt-4 and gpt-3.5-turbo) LLMs to generate personalized emails.
• Quality Checking: Uses gpt-4o to compare the two generated emails and select the best one.
• Orchestration with LangGraph: agents and tools are modelled using LangGraph (gpt-3.5-turbo).
• Interactive Notebooks: All project components are contained within Jupyter Notebooks for ease of use and experimentation.

WIP

• Email Optimization: Implement a basic MORL framework to adjust and optimize the email content iteratively based on feedback from previous email campaigns.

Installation

LeadGenAI uses Poetry for dependency management. Follow the instructions below to set up your environment.

Prerequisites

• Python 3.12
• Poetry (If not installed, you can find the installation instructions here)

Setup

1. Clone the repository:

   git clone https://github.com/mister-rao/LeadGenAI.git
   cd LeadGenAI

2. Install the dependencies using Poetry:

   poetry install

3. Activate the virtual environment:

   poetry shell

4. Launch Jupyter Notebook:

   jupyter lab

Usage

Once the environment is set up, you can open the provided Jupyter notebook to start generating personalized lead emails using the integrated LLMs.

License

This project is licensed under the GNU GENERAL PUBLIC LICENSE. See the LICENSE file for more details.

Acknowledgements

• LangGraph for orchestration support.
• OpenAI for providing powerful models for content generation.