TripTales

TripTales: AI-Powered Travel Planner & Visualizer

Comment 1

AWS & Impetus GenAI Hackathon Submission

✨ Use Case:

TripTales - GenAI Travel Planner

📅 Project Name:

TripTales AI Planner

✍️ Team Members:

  • Mohammad Iqbal (Solo Developer)

💡 Inspiration

The inspiration behind TripTales came from the frustration travelers face when planning trips: juggling hotel bookings, weather forecasts, visa checks, and creating a personalized itinerary — all from multiple sources.

"What if an AI assistant could handle all of it in one go — intelligently and visually?"

We wanted to harness Generative AI to create a complete travel experience planner that feels like having a personal travel concierge.

🔎 Problem Statement:

Travel planning is time-consuming and overwhelming, especially when travelers need to gather data from multiple sources — weather, hotels, visa, routes, and personalized preferences.

Scope: To automate the end-to-end travel planning process with GenAI — generating itineraries, images, hotel data, packing lists, visa info, and maps, all through a single app.

🚀 What It Does

TripTales is a GenAI-powered travel planner that transforms a few simple inputs into a complete, personalized travel experience — all in minutes. Here's what it does:

  • 📝 Collects User Preferences

    • Number of travelers (adults, children)
    • Destination, duration, budget
    • Start date, interests (nature, adventure, culture, etc.)
    • Image style preference (photo, sketch, painting, etc.)
    • Minimum hotel rating and search radius

  • 🌦️ Fetches Real-Time Weather

    • Uses Open-Meteo API to get daily and hourly forecasts
    • Adjusts itinerary suggestions accordingly

  • 🗺️ Suggests Hotels with Map Integration

    • Recommends hotels within a user-defined radius and rating
    • Displays exact locations on an interactive map

  • 📅 Generates a Smart Itinerary

    • Daily plans (Morning, Afternoon, Evening)
    • Weather- and interest-aware activities

  • 🖼️ Generates AI Images

    • Tourist destinations, food, packing visuals using Titan

  • 🍱 Suggests Local Food

    • AI-generated visuals for a must-try food list

  • 🧳 Provides a Packing List

    • Tailored to weather, duration, and destination

  • 🛂 Checks Visa Requirements

    • Offers general visa info based on nationality

  • 📄 Generates a Downloadable PDF

    • Consolidates all information in a shareable format

🛠️ How We Built It

Using Amazon Bedrock to connect:

  • Claude 3 Haiku for itinerary, food, visa, and packing generation
  • Titan for high-quality image generation
  • Streamlit for frontend UI
  • OpenWeather, Google Maps API for real-time data integration

🧗 Challenges We Faced

  • Structuring consistent output from Claude
  • Handling vague or missing location info for image prompts
  • Generating realistic AI visuals for diverse content types
  • Preventing latency and performance drops with real-time APIs
  • Ensuring session-based isolation to avoid data leakage

🏆 Accomplishments We're Proud Of

  • Fully functional end-to-end AI travel planner built within the hackathon window
  • Seamless integration of multiple AWS Bedrock foundation models
  • Real-time weather-aware itinerary generation
  • High-quality AI image generation and PDF export

📘 What We Learned

  • Integrating multiple foundation models via Amazon Bedrock
  • Crafting effective prompts for structured AI responses
  • Using AI outputs in a production-grade app
  • Optimizing AI and API balance for performance

🔮 What's Next for TripTales

  • Add login/personalization and user history
  • Integrate real-time transport APIs (flights, trains)
  • Support multi-city travel plans
  • Enable voice-based assistant interface
  • Provide offline itinerary access via mobile
  • Use enhanced Google Maps search and routes

📊 Tech Stack

  • Claude 3 Haiku, Amazon Titan
  • Python, Streamlit
  • Amazon Bedrock, EC2, VPC/IAM
  • MongoDB Atlas, OpenWeather, OpenRouteService

🧱 Architecture Overview

-Mermaid-based diagrams available in README.: https://github.com/miqbal303/triptales-genai-aws-bedrock/blob/main/README.md

Includes: Input → Caching → AI Generation → External APIs → Output (PDF)

💰 Cost Estimates

  • Bedrock (Claude + Titan): ~$0.25/user
  • EC2 (t3.medium): ~$30/month
  • MongoDB Atlas: Free tier

Estimated Monthly Cost: ~$30 for 1,000 users

🎯 Impact & Limitations

Impact:

  • 30% faster planning
  • 50% less manual work

Limitations:

  • Image resolution capped
  • No auto-updates for itinerary
  • Hotel data coverage may vary
  • Hotel Api work around ~15 call

🔐 Ethical & Security Considerations

  • No personal data storage beyond session
  • Claude prompt tested for fairness/bias
  • WCAG-compliant Streamlit UI
  • Transparent AI-generated content

🔗 Project Links

# Clone repository
git clone https://github.com/miqbal303/triptales-genai-aws-bedrock
cd triptales-genai-aws-bedrock

#Create .env file from GitHub secrets (replace with actual GitHub secrets or ensure secrets.env exists in the repo)
cp secrets.env .env
# Install dependencies
pip install -r requirements.txt

# Run application
streamlit run app.py

Built With

Share this project:

Updates