Set Up Credentials in Google Cloud Platform (GCP)

• Navigate to the Google Cloud Console.
• Create a new project or select an existing project.
• Enable the Vertex AI API for the project.
• Navigate to the "IAM & Admin" section and create a new service account with Vertex AI Admin role.
• Generate a JSON key for the service account and save it securely. This file will be used to authenticate your requests.

Prepare Your Bubble App

• Go to your Bubble app's editor, and open the Plugins tab.
• Install the "API Connector" plugin.
• Click the "Add another API" button and give your API a name like "Vertex AI".

Create an API Call in Bubble

• Under the "Vertex AI" API, click "Add a New Call" to create a new API call.
• Provide a name for the API call, such as "Generate Response".
• Set the method to POST.
• For the URL, use the Vertex AI endpoint for your region, e.g., https://-aiplatform.googleapis.com/v1/projects//locations//endpoints/:predict.
• In the "Headers" section, add the following headers:
  • Content-Type: application/json
  • Authorization: Bearer [INSERT_YOUR_ACCESS_TOKEN_HERE]
• In the "Body" section, use JSON formatting to create the request body required by Vertex AI. It might look something like this:

  {
    "instances": [{
      "input_data": ""
    }]
  }
      

Add Authentication Using the Service Account

• Go to the "Settings" tab in your Bubble editor.
• Scroll down to the "API Keys" section.
• Add a new key labeled "VertexAI\_ServiceAccountKey".
• Paste the contents of your JSON key file generated earlier.
• In the API Connector, use this key to fetch an OAuth token:
  • Set up another API call named "Get OAuth Token".
  • Method: POST
  • URL: https://oauth2.googleapis.com/token
  • Headers:
    • Content-Type: application/x-www-form-urlencoded
  • Body: Use a form-encoded body like:

    {
      "grant\_type": "urn:ietf:params:oauth:grant-type:jwt-bearer",
      "assertion": 
    }
            

Save and Test Your API Call

• Save all your settings and go back to your app's design interface.
• Set up a workflow to trigger the API call when an event occurs, like a button click.
• Use the response in your app's UI as needed.
• Test your configuration thoroughly to ensure everything works as expected.