# LLM Bill Chat App

This project is a proof of concept for a chat application utilizing a Large Language Model (LLM) to assist users with their telecom billing inquiries. The application is built using Python and Streamlit, providing an interactive web interface for users to engage with.

## Features

- Maintain chat conversation context (history)
- Allow users to query their billing information
- Compare the last bills and provide insights
- Respond exclusively with the user's own billing information
- Augment the prompt instructions with user's text recognized entities - NER -> Dynamic prompting# Aplicația LLM Bill Chat

## Project Structure

hf-llm-bill-chat/
│
├── app.py
├── pages/
│   ├── One_model.py
│   ├── Two_models.py
│
├── data/
│   └── user_data/
│       └── user_data_724077190.json
└── logs/
    └── conversation_logs.json


## Code Implementation - Main Script

- Initialize Streamlit app
- Load user data and display existing bills
- Handle file uploads for new bills
- Maintain chat conversation context
- Leverage user text entities recognition (NER) for dynamic prompting
- Call OpenAI models (GPT-4o and GPT-4o-mini) for responses
- Display responses in two parallel columns
- Log conversation details

## Future Improvements

- Enhance Natural Language Processing (NLP) capabilities for better entity recognition
- Implement user authentication for secure access to billing information
- Add support for more telecom providers and bill formats
- Improve the user interface for better user experience
- Integrate with payment gateways for bill payments

## Conclusion

- The LLM Bill Chat App demonstrates the potential of using Large Language Models for telecom billing assistance.
- The app provides a user-friendly interface for querying billing information and getting insights.
- Future improvements can enhance the app's capabilities and user experience.