Skip to content

AI TravelOps is an end-to-end, LLM-powered trip planning system that generates complete travel itineraries for any city worldwide using real-time data. The system integrates weather forecasting, attraction discovery, hotel cost estimation, currency conversion, expense breakdown, and summary generation—fully automated through an LLMOps pipeline.

License

Notifications You must be signed in to change notification settings

PRIYANGSHU018/AI-TravelOPs-Autonomous-AI-Travel-Planner-End-to-End-Agentic-LLMOPs-Pipeline

Repository files navigation

🌍 AI-TravelOps — Autonomous Travel Planner with Real-Time Tools & LLM Orchestration

A fully automated, tool-using travel planning agent with real-time weather, attractions, restaurants, flights, hotel data, currency conversion, and cost estimation.


🚀 Overview

AI-TravelOps is an intelligent, agent-based travel planning system built with:

  • LangGraph for agent orchestration
  • LangChain ToolNodes for real-time API calls
  • Groq / OpenAI models for LLM reasoning
  • Foursquare / Tavily / Serper / Open-Meteo for global live data
  • FastAPI / Streamlit frontend (optional)
  • Docker + EC2 support for deployment

The system automatically builds multi-day itineraries, fetches real-time flight prices, weather forecasts, hotel listings, restaurants, transportation options, and produces a cost-optimized travel plan.


Key Features

🔎 Autonomous Agent Workflow

  • Uses LangGraph to manage multi-step tool calls
  • Zero hallucinations — ONLY uses real API results
  • Multi-tool data fusion

🌦️ Real-Time Weather

  • 100% tool-based
  • Multi-day forecast (N days)
  • Accurate min–max temperatures + conditions

✈️ Live Flight Prices

  • Uses flight API tool

  • Returns:

    • airline
    • currency
    • price
    • INR converted price

🏨 Hotel Search

  • Powered by Tavily (search) and Foursquare (places)

  • Returns:

    • hotel name
    • address
    • category
    • coordinates
  • No hallucinated hotel prices.

🥘 Restaurants & Attractions

  • Real-time data from Foursquare
  • Includes category, address & geo-coordinates

🚌 Transportation Mapping

  • Metro, bus, taxi, train, airport access

💰 Cost Calculator

Using internal tools:

  • estimate hotel cost
  • calculate total trip cost
  • daily budget calculation
  • full INR conversion

📅 Auto-Generated Multi-Day Itinerary

For N days:

  • Morning / Afternoon / Evening activity suggestions
  • Always uses only tool-returned places

🧠 Architecture

User Query
     ↓
AI-TravelOps Agent  (LangGraph)
     ↓
 ┌────────── Tools Layer ──────────┐
 │ • WeatherInfoTool                │
 │ • PlaceSearchTool               │
 │ • HotelSearchTool (Tavily)      │
 │ • FlightCostTool                │
 │ • CalculatorTool                │
 │ • CurrencyConverterTool         │
 └──────────────────────────────────┘
     ↓
Structured Travel Plan (Markdown)

🛠️ Tech Stack

Backend

  • Python 3.10+
  • FastAPI & Streamlit (optional UI)
  • LangChain
  • LangGraph
  • Groq LLaMA / Mixtral / OpenAI models

Tools & APIs

  • Foursquare Places API
  • Tavily Search API
  • Serper Search API
  • Open-Meteo Weather API
  • Custom Flight Pricing API
  • Currency conversion tool

Deployment

  • Docker
  • EC2 Ubuntu Server
  • Nginx reverse proxy (optional)

📂 Project Structure

AI-TravelOPs/
│
├── agent/
│   ├── __init__.py
│   └── agentic_workflow.py
│
├── config/
│   ├── __init__.py
│   └── config.yaml
│
├── tools/
│   ├── __init__.py
│   ├── weather_info_tool.py
│   ├── place_search_tool.py
│   ├── hotel_search_tool.py
│   ├── flight_cost_tool.py
│   ├── expense_calculator_tool.py
│   ├── currency_conversion_tool.py
│   ├── arthamatic_op_tool.py              # (Your helper tool)
│
├── utils/
│   ├── __init__.py
│   ├── model_loader.py
│   ├── config_loader.py
│   ├── expense_calculator.py
│   ├── weather_info.py
│   ├── save_to_document.py
│
├── prompt_library/
│   ├── __init__.py
│   └── prompt.py
│
├── test/
│   ├── test_env.py
│   ├── test_currency.py
│   ├── test_weather.py
│   ├── test_place_search.py
│   ├── test_place_debug.py
│   ├── test_flight.py
│   ├── test_hotel.py
│   ├── test_groq_bind.py
│   ├── test_fsq.py
│   ├── testenwea.py
│   └── test-raw.py
│
├── output/
│   └── (generated travel plans)
│
├── streamlit_app.py       # Your UI (from screenshot)
├── main.py                # Entry point (CLI / LangGraph runner)
│
├── Dockerfile
├── requirements.txt
├── pyproject.toml
├── setup.py
├── LICENSE
├── README.md
├── question.txt
└── .env


⚙️ Setup Instructions

1️⃣ Clone Repository

git clone https://github.com/<your-username>/AI-TravelOps.git
cd AI-TravelOps

2️⃣ Create Environment

python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

3️⃣ Add .env File

FOURSQUARE_API_KEY=xxxx
TAVILY_API_KEY=xxxx
SERPER_API_KEY=xxxx
FLIGHT_API_KEY=xxxx
GROQ_API_KEY=xxxx
OPENWEATHER_API_KEY=xxxx
EXCHANGE_RATE_API_KEY=xxxx
GEOAPIFY_API_KEY=xxxx

4️⃣ Run (backend)

python main.py

🐳 Docker Deployment (Production Ready)

Build

docker build -t travelops .

Run

docker run -d -p 8000:8000 --env-file .env travelops

Access at:

http://<EC2-Public-IP>:8000

🤝 Contributing

Pull requests are welcome! Submit issues / feature requests in the "Issues" tab.


📜 License

MIT License.


⭐ If you like this project, star the repo!

Your support motivates future improvements ❤️


About

AI TravelOps is an end-to-end, LLM-powered trip planning system that generates complete travel itineraries for any city worldwide using real-time data. The system integrates weather forecasting, attraction discovery, hotel cost estimation, currency conversion, expense breakdown, and summary generation—fully automated through an LLMOps pipeline.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages