Skip to main content
Version: 0.0.38

AWS Bedrock

Connecting AWS Bedrock with Lakehousecat allows users to utilize AWS's robust capabilities for Provider Models efficiently. This guide will explain how to configure an AWS Bedrock connection within the Lakehousecat workspace.

Prerequisites​

Before configuring AWS Bedrock, ensure you have:

  • Access to Lakehousecat as an administrator or model manager
  • AWS account credentials (AWS Access Key ID, AWS Secret Access Key)
  • Understanding of AWS region specifications

Step-by-Step Guide​

  1. Enter the Workspace

    • Log into your Lakehousecat workspace to begin the Provider Models setup.

  2. Navigate to Models

    • Select the 'Models' menu to access the Provider Models area.

  3. Select Provider Models

    • Click on 'Provider Models', then press the '+' icon to begin adding new provider configurations.

  4. Choose AWS Provider

    • In the provider selection combobox, select 'AWS' to proceed with the configuration.

Step 3: Configure AWS Bedrock​

  • Configuration Name: Enter a meaningful name for your AWS Bedrock connection. Adopting a consistent naming convention is recommended, especially with multiple providers.

  • Model Types: Choose from various models available, such as Chat, Embedding, Text-to-Speech, and Speech-to-Text.

  • AWS Access Key ID: Provide your AWS Access Key ID for authentication purposes.

  • AWS Secret Access Key: Input your AWS Secret Access Key to complete the credential setup.

  • Region: Specify the AWS region where your infrastructure and services are hosted.

  • Inference Profile: Retrieve the inference profile arm using the AWS Backdrop List Inference Profiles command via AWS CLI. Ensure authentication with the AWS Cloud Provider to access the profile arms.

Refer to AWS documentation for detailed access setup if needed.

Step 4: Validate and Set Up the Connection​

  1. After filling out the necessary information, click the Create button to initiate connection validation.
  2. Lakehousecat will verify the entered parameters. Make sure all details are correct to prevent validation errors.
  3. Upon successful validation, your AWS Bedrock connection will be established and ready for use.

Best Practices​

  • Consistent Naming: Develop a systematic naming convention for your configuration names to enhance clarity and management.

Troubleshooting​

If issues arise during connection setup:

  • Credential Errors: Double-check AWS Access Key ID and AWS Secret Access Key for accuracy.
  • Region Mismatch: Ensure the specified region corresponds to your AWS settings.
  • Profile Inference Retrieval: Verify CLI authentication with AWS for inference profiles access.