TravelForge ✈️⚒️

TravelForge is an AI-powered travel itinerary generator designed to create personalized travel plans tailored to your unique preferences. TravelForge leverages multi-agent workflows to streamline the travel planning process, ensuring you get a customized and enjoyable travel experience.

🤖 Agents Overview

We use the LangGraph Library to design our agentic workflow. The following agents work together to create a seamless and personalized travel itinerary:

1. Researcer: Gathers information about destinations, attractions, accommodations, and more.
2. Recommender: Suggests activities and accommodations options based on user preferences and interests upon recieving the research results.
3. Itinerary Generator: Creates a detailed itinerary based on the user's preferences and the recommendations provided by the Recommender.
4. Itinerary Formatter: Formats the API's response for presentation and delivery to the user.
5. (WIP) Feedback Agent: Collects user feedback on the itinerary and makes adjustments based on the feedback.
6. (WIP) Mapper: Generates an interactive map of all the sites and activities in the itinerary for the user to explore.

🎥 Demo

https://www.loom.com/share/bd4506ed6f394c45886ec034a04c2cda?sid=94765c4d-fd73-4c63-b33e-5993d3b4f613

🌟 Features

• Personalized Itineraries: Create travel plans that align with your interests and preferences.
• Comprehensive Planning: Includes details on attractions, accommodations, dining, and activities.
• Engaging Design: Enjoy a visually appealing and well-organized itinerary layout with expandable cards.
• Quality Assurance: Iterative feedback and refinement ensure a satisfying travel plan.
• User-Friendly Interface: Easy-to-use platform for setting preferences and generating itineraries.

🛠️ How It Works

1. Setting Preferences: Users input their travel interests, preferred destinations, and travel dates.
2. Automated Curation: The Research Agent gathers information on destinations, attractions, accommodations, and activities.
3. Itinerary Creation: The Recommender Agent suggests activities and accommodations based on user preferences, and the Itinerary Generator creates a detailed itinerary.
4. Delivery: The Itinerary Formatter formats the itinerary for presentation and delivery to the user.

🗂️ Installation

You will need API keys from both Tavily and OpenAI to run the app. You can sign up for a free account on their respective websites to get your API keys.

1. Clone the repo

   git clone https://github.com/taimurshaikh/travelforge.git

2. Export your API Keys

   export TAVILY_API_KEY=<YOUR_TAVILY_API_KEY>
   export OPENAI_API_KEY=<YOUR_OPENAI_API_KEY>

   ...OR add them to an .env file in the root directory

    TAVILY_API_KEY=<YOUR_TAVILY_API_KEY>
    OPENAI_API_KEY=<YOUR_OPENAI_API_KEY>

3. Install required packages for both the frontend and backend

   cd frontend
   npm install
   cd ../backend
   python -m venv venv && source venv/bin/activate
   pip install -r requirements.txt

4. The simplest way to start the app is to run the shell script in the root directory. First, ensure the script is executable:

   chmod +x startup-script.sh

   Then run the script:

   ./startup-script.sh

5. The app should now be running on http://localhost:5173.
6. Start planning your next adventure ✈️ ✈️ ✈️! Note: You should exit the app by pressing Ctrl+D in the terminal to stop the server and kill both frontend and backend tasks.

🛣️ Roadmap

Checklist of features and functionalities that are planned for future releases. These will be turned into GitHub milestones in the future.

• Docker Support: Finish dockerizing the application for easy deployment and scaling.
• Feedback Agent: Collect user feedback on the itinerary and make adjustments based on the feedback.
• Mapper: Generate an interactive map of all the sites and activities in the itinerary for the user to explore.
• Multi-User Support: Allow multiple users to create and manage their travel itineraries.
• Collaborative Planning: Enable users to share and collaborate on travel plans with friends and family.
• Mobile App: Look into developing a mobile app version of TravelForge for on-the-go travel planning.

❗Known Issues

These will be migrated to the issues tab in the future.

• Docker Support: Docker support is still a work in progress. There is an open bug with the frontend Dockerfile not including the packages for building to all targets (see this issue)that needs to be resolved.
• Accomodation Images: The current algorithm for matching the images returned by the Researcher's Tavily Request to the accomodations is a heuristic, and sometimes leads to incorrect images being displayed for a given accomodation.
• Days with No Activities: Depending on the OpenAI call, there may be days where no activities are suggested. This is a known issue and will be addressed in future releases.

🤝 Contributing

Contributing could be as simple as opening an issue to report a bug or request a new feature. If you would like to contribute to the codebase, please make a fork and open a pull request with your changes. We would love to see your contributions! As the project scales, we will be adding more detailed contribution guidelines.

📩 Contact Us

For support or inquiries, please reach out to us:

• Email

Join us in revolutionizing the travel planning experience with TravelForge!

🙏 Sources and Acknowledgements

Tavily OpenAI LangGraph LangChain

We would like to sincerely thank the team at Tavily for providing inspiration for this project as part of their take-home challenge.