🌍 AI Travel Agent (LangGraph Edition)

A fully functional Agentic AI Travel Planner built using LangGraph, OpenAI, and SingleStore.

This AI system helps users:

• ✔️ Get hotel recommendations
• ✔️ Find flights
• ✔️ Fetch weather information
• ✔️ Cache results in SingleStore. Get your free SingleStore account
• ✔️ Build reproducible agentic workflows

All powered by a modular LangGraph graph with independent agents.

---

Live demo: https://travel-agentic-ai.onrender.com/

🚀 Features

For a detailed description of the agent architecture and step-by-step examples, see docs/agent_architecture.md.

📦 Installation

Clone the repository:

git clone https://github.com/pavanbelagatti/Agentic-AI-Travel-Agent.git
cd Agentic-AI-Travel-Agent.git

Create a virtual environment

python3 -m venv venv

source .venv/bin/activate     # For Mac/Linux
OR
.\.venv\Scripts\activate      # For Windows

Install dependencies

pip install -r requirements.txt

🔧 Environment Variables

Copy .env.example to .env and fill in your real values locally. Do NOT commit your .env file to the repository.

cp .env.example .env
# then edit .env and add your secrets locally

If you are deploying to a platform (Render, Streamlit Cloud, etc.) add the same variables to the platform's secret manager or environment settings.

🗄️ Setting Up SingleStore

• Signup & Log in to SingleStore Account
• Click Create Database → Name it TravelAG or whatever you wish to name it
• Open the SQL Editor in the SingleStore dashboard
• Run the following table creation queries:

Accommodations table

CREATE TABLE IF NOT EXISTS accommodations (
    id BIGINT AUTO_INCREMENT PRIMARY KEY,
    provider VARCHAR(100),
    provider_item_id VARCHAR(200),
    name VARCHAR(255),
    location_city VARCHAR(255),
    location_country VARCHAR(255),
    bedrooms INT,
    price_per_night DOUBLE,
    rating DOUBLE,
    url TEXT,
    vector BLOB,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

Flights table

CREATE TABLE IF NOT EXISTS flights (
    id BIGINT AUTO_INCREMENT PRIMARY KEY,
    provider VARCHAR(100),
    airline VARCHAR(200),
    origin VARCHAR(10),
    destination VARCHAR(10),
    depart_time VARCHAR(50),
    arrive_time VARCHAR(50),
    price DOUBLE,
    url TEXT,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

These tables will remain empty initially — your LangGraph agents will store results into them automatically during the first run.

Running the application

python main.py

You’ll be asked:

• Origin airport/city
• Destination
• Start & end dates
• Bedrooms
• Budget
• Minimum rating

After the LangGraph workflow completes, you’ll get:

• Weather summary
• Top hotel recommendations
• Top flight options