AI Personal Assistant

This project is an AI-powered personal assistant built using Streamlit for the frontend and FastAPI for the backend. The backend interacts with the OpenAI API to generate responses, and the frontend provides a user-friendly chat interface.

Features

• Chat Interface: Users can interact with the AI assistant via a chat interface.
• Session Management: Each conversation is associated with a unique session ID.
• Conversation History: The chat history is stored in an SQLite database.
• OpenAI Integration: The backend uses the OpenAI API to generate responses.

---

Prerequisites

Before running the project, ensure you have the following installed:

1. Docker: Install Docker
2. Docker Compose: Install Docker Compose

---

Project Structure

ai-personal-assistant/
├── backend/
│   ├── app.py/                  # Backend application code
│   ├── Dockerfile            # Dockerfile for the backend
├── frontend/
│   ├── streamlit_app.py/                  # Frontend application code
│   ├── Dockerfile            # Dockerfile for the frontend
├── docker-compose.yml        # Docker Compose configuration
└── README.md                 # Project documentation

---

Setup and Running the Project

Step 1: Clone the Repository

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

Step 2: Build and Run the Services

Use Docker Compose to build and run the backend and frontend services:

docker-compose up --build

This will:

1. Build the Docker images for the backend and frontend.
2. Start the backend service on port 8000.
3. Start the frontend service on port 8501.

Step 3: Access the Application

• Frontend: Open your browser and go to http://localhost:8501.
• Backend: The backend API will be available at http://localhost:8000.

---

Using the Application

1. Enter OpenAI API Key:

   • On the left sidebar of the Streamlit frontend, enter your OpenAI API key.

2. Start Chatting:

   • Type your message in the chat input box and press Enter.
   • The AI assistant will respond to your queries.

3. Reset Conversation:

   • Use the "Start New Conversation" button in the sidebar to reset the chat history.

---

Docker Compose Configuration

The docker-compose.yml file defines two services:

1. Backend:

   • Built from the ./backend/Dockerfile.
   • Exposes port 8000 for the FastAPI backend.

2. Frontend:

   • Built from the ./frontend/Dockerfile.
   • Exposes port 8501 for the Streamlit frontend.
   • Depends on the backend service.

---

Backend Details

The backend is built using FastAPI and provides the following functionality:

• API Endpoint: POST /llm

  • Accepts a JSON payload with the user's question, conversation history, session ID, and OpenAI API key.
  • Returns the AI-generated response.

• Database:

  • Conversations are stored in an SQLite database (conversations.db).

---

Frontend Details

The frontend is built using Streamlit and provides the following functionality:

• Chat Interface:

  • Users can enter their OpenAI API key in the sidebar.
  • Users can interact with the AI assistant via a chat interface.

• Session Management:

  • Each conversation is associated with a unique session ID.

---

Stopping the Services

To stop the services, run:

docker-compose down

---

Troubleshooting

1. Docker Compose Fails to Start:

   • Ensure Docker and Docker Compose are installed correctly.
   • Check the logs for errors using docker-compose logs.

2. Frontend Cannot Connect to Backend:

   • Ensure the backend service is running and accessible at http://localhost:8000.
   • Verify the frontend is correctly configured to call the backend API.

3. OpenAI API Key Not Working:

   • Ensure the API key is valid and has sufficient credits.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

Acknowledgments

• OpenAI: For providing the GPT-3.5-turbo model.
• Streamlit: For the easy-to-use frontend framework.
• FastAPI: For the high-performance backend framework.

---

Feel free to customize this README.md to better suit your project. Let me know if you need further assistance!