This project implements a text-based developer assistant using a Mixture of Experts (MoE) model. The assistant consists of a prime agent and several secondary agents specialized in different tasks such as code writing, code debugging, and code optimization.
agent.py
: Defines the PrimeAgent and SecondaryAgent classes.app.py
: Main entry point for running the Gradio interface.network.py
: Contains functions to create neural network models for text processing and the gating network.requirements.txt
: Lists the dependencies required for the project.train.py
: Implements the training and evaluation pipeline.interface.py
: Sets up the Gradio interface for the chat window and code display.README.md
: Project documentation and setup instructions.Follow these steps to set up the project on your local machine:
git clone https://github.com/Dnnsdesigns/AgenticDeveloper.git
cd AgenticDeveloper
python -m venv venv
source venv/bin/activate # On Windows use `venv\Scripts\activate`
pip install -r requirements.txt
python app.py
The AgenticDeveloper provides an interactive Gradio interface with two main components:
To use the assistant, enter your request in the chat window and the assistant will respond with suggestions or code snippets based on the input.
We welcome contributions to the AgenticDeveloper project! Here are some guidelines to help you get started:
git clone https://github.com/YOUR_USERNAME/AgenticDeveloper.git
cd AgenticDeveloper
git checkout -b feature/your-feature-name
git commit -m "Add feature/your-feature-name"
git push origin feature/your-feature-name
Here are some potential future improvements and features for the AgenticDeveloper: