# Multi-Agent Hospitality AI Assistant

An AI-powered restaurant assistant that answers customer questions, searches menu data, and manages reservation workflows through a multi-agent architecture.

## Features

* Answers questions about restaurant information and policies

* Searches menu items by category, price, ingredients and preferences

* Creates, looks up, modifies and cancels reservations

* Handles callback requests

* Routes requests to specialised agents

* Uses separate RAG pipelines for restaurant information and menu search

* Generates grounded responses using retrieved context

* Stores and manages reservation data through Supabase

## Architecture

The application uses an orchestrator to identify the user’s intent and route the request to the appropriate agent:

* info\_agent.py — restaurant information queries

* menu\_agent.py — menu search and filtering

* reservation\_agent.py — reservation and callback operations

* orchestrator.py — intent detection and agent routing

## Tech Stack

* Python

* Retrieval-Augmented Generation

* Pinecone

* Groq LLM

* SentenceTransformers

* LangChain

* Supabase

* Pandas

* PDF and CSV processing

* Streamlit

## Project Structure

* agents/ — specialised information, menu and reservation agents

* app.py — application interface

* orchestrator.py — request routing logic

* info\_restaurant.py — restaurant-information RAG pipeline

* menu.py — menu retrieval and filtering

* reservation.py — reservation workflow

* config.py — environment-based configuration

* Fake restaurant info.pdf — synthetic restaurant knowledge base

* saffron\_table\_menu\_dataset\_ffs\_bb.csv — synthetic menu dataset

## Setup

Install the dependencies:

pip install -r requirements.txt

Create a .env file using .env.example and add the required credentials:

GROQ\_API\_KEY=your\_groq\_api\_key

PINECONE\_API\_KEY=your\_pinecone\_api\_key

SUPABASE\_URL=your\_supabase\_url

SUPABASE\_KEY=your\_supabase\_key

Run the application:

streamlit run app.py

## Security

Credentials are loaded through environment variables and are not stored in the repository.

## Note

The restaurant information and menu data included in this repository are synthetic and intended for demonstration purposes.