🌍 LuxeTravel AI — Intelligent Multi-Agent Travel Concierge

Google AI Agents Intensive — Capstone Project (Fall 2025)

---

📖 Overview

LuxeTravel AI is a powerful multi-agent travel planning system that automates research, hotel finding, and itinerary creation.
Instead of manually browsing dozens of websites, LuxeTravel’s agents research, curate, and deliver a complete itinerary in seconds.

Designed for the Concierge Agents Track, this project demonstrates:

• Multi-Agent Orchestration
• Tool-Augmented Agents (Web Search)
• Google Gemini Reasoning
• Session Memory
• Real-time Observability in Streamlit

---

🤖 Multi-Agent Team

1. 🕵️ Researcher Agent — “The Scout”

Tools: DuckDuckGo Search
Tasks:

• Live weather
• Travel advisories
• Events & festivals
• Cultural insights

---

2. 🏨 Hotel & Dining Agent — “The Concierge”

Tools: DuckDuckGo Search
Tasks:

• Hotels based on budget
• Dining recommendations
• Price comparison
• Ratings-based filtering

---

3. 📅 Planner Agent — “The Architect”

Tools: Pure LLM
Tasks:

• Day-by-day itinerary
• Cost estimation
• Travel flow optimization
• Final trip summary

---

🏗️ Architecture

graph TD
    A[User Input] --> B[Streamlit UI]
    B --> C[🕵️ Researcher Agent]
    C --> M1[Session Memory]

    M1 --> D[🏨 Hotel Agent]
    D --> M2[Session Memory]

    M2 --> E[📅 Planner Agent]
    E --> B[Final Itinerary]

Loading

Core Architecture Features

• Live data via Web Search
• Sequential multi-agent pipeline
• Session-based memory
• Real-time status updates
• Gemini-powered synthesis

---

🛠️ Tech Stack

Component	Technology
Frontend UI	Streamlit
Multi-Agent Framework	Agno (Phidata)
AI Model	Google Gemini 1.5 Flash
Tools	DuckDuckGo Web Search
Environment	Python 3.11+

---

🚀 Installation & Setup

1. Clone Repo

git clone [https://github.com/Kiruthiyan/Travel-Planner-AI-Agent]
cd Travel-Planner-AI-Agent

2. Create Virtual Environment

python -m venv .venv

# Windows
.venv\Scripts\activate

# macOS / Linux
source .venv/bin/activate

3. Install Requirements

pip install -r requirements.txt

4. Add API Key

Create a .env file:

GEMINI_API_KEY=your_google_api_key_here

Get your free key from Google AI Studio.

5. Run App

streamlit run app.py

---

💡 How to Use

1. Choose destination, dates, budget, and interests

2. Click "Design My Perfect Trip"

3. Agents run sequentially:

   • Researcher
   • Hotel Finder
   • Planner

4. View results in:

   • Itinerary
   • Research
   • Hotels & Dining

---

🏆 Capstone Requirements Checklist

Requirement	Status
Multi-Agent System	✅ Done
Tools (Search, Custom)	✅ DuckDuckGo Web Search
Gemini Model	✅ 1.5 Flash
Memory	✅ Session Memory
Observability	✅ UI Live Status
Sequential / Parallel / Scoped Agents	✅ Sequential Pipeline
Agent Evaluation	✅ Done
A2A Protocol	✅ Done
Deployment	✅ Done Deploy to Streamlit Cloud

---

📦 requirements.txt

streamlit
agno
google-generativeai
duckduckgo-search
python-dotenv
tenacity

---

🤝 License

MIT License. Built with ❤️ for the Google AI Agents Intensive Capstone Project.

---