🚀 AI-Powered Code Assistant

This project is an AI-powered code assistant that leverages cutting-edge natural language processing and retrieval-augmented generation (RAG) techniques. The assistant provides intelligent code assistance, document parsing, and interaction with OpenAI's powerful language models, such as GPT-4, to deliver accurate and context-aware code suggestions and analysis.

📑 Table of Contents

• Features
• Installation
• Usage
• Project Structure
• Common Utilities
• FastAPI Module
• LlamaParse Utilities
• OpenAI Utilities
• Contributing
• License

🌟 Features

• ⚡ FastAPI Backend: A robust and lightweight API framework that handles code-related queries and integrates easily with the AI models.
• 📄 LlamaParse Integration: Advanced document parsing using LlamaParse for structured information extraction from complex documents (e.g., financial reports, PDFs).
• 🤖 OpenAI GPT-4 Integration: Seamless integration with OpenAI’s GPT-4 for intelligent code generation, code analysis, and language understanding.
• 🖥️ Streamlit-Based User Interface: A user-friendly web interface that allows users to interact with the AI-powered code assistant effortlessly.
• 🛠️ Support for Multiple Programming Languages: Provides coding assistance across a variety of programming languages, including Python, JavaScript, Go, and more.

🛠 Installation

1. Clone the repository:

   git clone https://github.com/your-username/ai-code-assistant.git
   cd ai-code-assistant

2. Install the required dependencies for FastAPI, LlamaParse, and OpenAI integrations:

   pip install -r requirements.txt

3. Set up your environment variables: Create a .env file in the root directory and add the following variables:

   OPENAI_API_KEY=your_openai_api_key
   LLAMA_CLOUD_API_KEY=your_llama_cloud_api_key

   These keys are required for interacting with the OpenAI and LlamaParse APIs.

🚀 Usage

1. Start the FastAPI server:

   uvicorn fastapi.main:app --reload

   This will start the backend service, which handles requests and provides the AI-powered code assistant's functionality.

2. Run the Streamlit application:

   streamlit run common/examples/app-streamlit.py

   The Streamlit app provides a web-based interface where users can interact with the assistant. After launching, access the app at http://localhost:8501.

3. Access the FastAPI docs: Visit http://localhost:8000/docs to explore the API endpoints and try out requests using FastAPI's built-in documentation system.

🗂️ Project Structure

GEN-AI-UTILS/
│
├── common/               # Utility scripts for shared functionality
├── fastapi/              # FastAPI backend application code
├── llama-parse/          # Scripts and notebooks for document parsing using LlamaParse
├── openai/               # Utilities for interacting with OpenAI's GPT-4 and vector stores
└── .gitignore            # Git ignore file for excluding files from version control

📂 Common Utilities

The common folder contains helper scripts that are used across various components of the project, such as user interface customization, messaging utilities, and OpenAI interaction.

• app-streamlit.py: Manages the layout and interactions of the Streamlit-based web application.
• custom_css_banner.py: Injects custom CSS to modify the web interface's banner.
• custom_css_main_page.py: Handles CSS injection for customizing the main page of the Streamlit app.
• message_utils.py: Helper functions for managing and formatting user-facing messages.
• openai_utils.py: Provides utility functions for interfacing with OpenAI’s GPT-4 API.

📂 FastAPI Module

The fastapi folder is responsible for managing the backend API of the project, handling incoming requests, and routing them to the appropriate AI models or utilities.

• main.py: Contains the FastAPI application with all the defined routes, dependencies, and request handlers.
• Procfile: Specifies how the FastAPI application should be run in production environments (e.g., using Uvicorn).
• requirements.txt: Lists all Python dependencies required for running the FastAPI server and other components.

📂 LlamaParse Utilities

The llama-parse folder holds scripts and tools for document parsing, particularly using the LlamaParse framework. It can process complex PDFs and extract structured data, such as financial statements.

• 01-advanced-RAG-with-LlamaParse.ipynb: A Jupyter notebook demonstrating the use of LlamaParse with a retrieval-augmented generation (RAG) model.
• 02-llama-parse-script-v2.py: A Python script for parsing documents and extracting information into markdown format.
• NVIDIA-10-Q-AUG.pdf: Sample document used for testing LlamaParse’s parsing capabilities.

📂 OpenAI Utilities

The openai folder contains utilities for interacting with OpenAI’s API and managing vector stores for document retrieval tasks.

• batch-update-vector-store-files.ipynb: A notebook to update vector stores in bulk for RAG or similar use cases.
• openai_assistant_respose.py: A script for managing the interaction between OpenAI’s GPT-4 and the user interface, handling responses from the assistant.

🤝 Contributing

We welcome contributions to the project! If you’d like to contribute, please fork the repository, make your changes, and submit a pull request. Be sure to follow the code style guidelines and provide adequate documentation.

📜 License

This project is licensed under the MIT License. Feel free to use, modify, and distribute this project under the terms of the license.