Google Updates BigQuery with New Pipelines, Version Control, and Claude AI Integration

Google has rolled out several significant updates to its BigQuery service, including feature renaming, version control capabilities, and new AI model integration options.

BigQuery Workflows Renamed to Pipelines

Google has renamed BigQuery workflows to BigQuery pipelines in the Google Cloud console. Users looking for more details can reference the introduction to BigQuery pipelines documentation for comprehensive information about this change.

New Version Control Capabilities in Preview

Google has introduced repositories and workspaces to BigQuery, enabling version control functionality. This new preview feature allows users to:

• Use repositories to perform version control on files using Git to record changes and manage file versions
• Utilize workspaces within repositories to edit code stored in the repository
• Configure repositories to either use Git directly on BigQuery or connect to third-party Git providers

This enhancement brings professional development practices directly into the BigQuery environment, though it remains in preview status.

Claude AI Integration Now Generally Available

BigQuery ML now supports creating remote models based on the Anthropic Claude model in Vertex AI. This fully released feature enables users to:

• Use the ML.GENERATE_TEXT function with these remote models to perform generative natural language tasks for text stored in BigQuery tables
• Evaluate Claude models using the ML.EVALUATE function

Google has made this capability generally available (GA), allowing users to implement advanced AI text generation directly within their BigQuery environment. Those interested can explore a dedicated tutorial on generating text using the ML.GENERATE_TEXT function.