An AI-powered travel planning assistant built with Java, Spring Boot. It orchestrates a sequential multi-agent workflow to search for flights, recommend hotels, and construct a complete day-by-day itinerary.
To use the underlying Gemini model (via Vertex AI), you need to set up a GCP project, configure your local authentication, and enable the Agent Platform API.
You can create a new GCP project using either the Google Cloud Console
- Visit the Google Cloud Console.
- Click the project dropdown in the top navigation bar and select New Project.
- Enter your project name/ID and click Create.
Once your project is created, configure your local environment to connect to it.
-
Set your active project:
gcloud config set project YOUR_PROJECT_ID -
Authenticate Application Default Credentials (ADC): This enables Vertex AI clients running on your local machine to authenticate automatically.
gcloud auth application-default login
-
Set the billing/quota project: This prevents API quota errors by linking calls to your billing account project.
gcloud auth application-default set-quota-project YOUR_PROJECT_ID
The SDK utilizes the Gemini model through the Vertex AI APIs (aiplatform.googleapis.com).
- Navigate directly to the Vertex AI API Overview (replace
YOUR_PROJECT_IDin the URL). - Click the Enable button.
Once GCP is set up, follow these steps to build and run the application.
- Java 17 or higher
- Docker (optional)
Compile the source code and package it into an executable JAR:
./gradlew build -x testStart the application with standard local configurations:
./gradlew bootRunThe server will start, by default listening on port 8080.
Open your web browser and navigate to:
http://localhost:8080
This opens the built-in interactive web UI where you can chat with the travel agent.