Manual pipeline maintenance can drain your team’s creativity and slow down innovation. Google’s new Data Engineering Agent in BigQuery, now in preview, aims to automate routine data engineering tasks, empowering teams to focus on strategic projects rather than daily upkeep. By leveraging advanced AI, this tool promises smarter, faster, and more autonomous data workflows.
Revolutionizing Pipeline Development and Maintenance
The Data Engineering Agent transforms how you design and manage data pipelines. Instead of writing complex code, you simply describe your requirements in natural language, and the agent generates robust SQL pipelines following best practices. Updating pipelines is just as easy, simply state your desired changes and the agent optimizes and updates the code, eliminating redundancy and improving efficiency.
- Natural language pipeline creation: Turn plain English instructions into fully functional SQL pipelines.
- Effortless pipeline modification: Adjust or optimize workflows with simple directives.
- Metadata enrichment: Integrate business glossaries and table documentation through Dataplex Universal Catalog.
- Custom business logic: Incorporate unique rules and user-defined functions seamlessly.
- Automated documentation: Generate comprehensive and clear documentation for every pipeline.

Accelerating Data Preparation and Modeling
Data preparation is often the most time-consuming phase of analytics. The Data Engineering Agent automates ingestion, deduplication, standardization, and sensitive data identification.
With Dataplex integration, it enforces data quality rules and encrypts columns containing personally identifiable information (PII), ensuring compliance and consistency throughout your workflows.
- Prepares data from Google Cloud Storage and multiple sources
- Applies data quality and formatting standards automatically
- Generates complex transformation code, including joins and aggregations
- Supports advanced modeling: create Data Vault or Star Schemas using natural language prompts
Early adopters like PRISA Media have used the agent to automate sophisticated modeling, such as managing slowly changing dimensions, highlighting its potential for complex enterprise use cases.
Smarter, Proactive Pipeline Troubleshooting
When a pipeline fails, downtime can quickly add up. Integrated with Gemini Cloud Assist, the Data Engineering Agent analyzes execution logs, identifies the root cause, and suggests actionable fixes. This acceleration of troubleshooting helps teams resolve issues quickly and keeps data workflows running smoothly.
- Rapid diagnosis and solution suggestions for failures
- Minimizes manual log analysis
- Speeds up recovery of stalled pipelines
Simplifying Migration and Modernization
Migrating to Google Cloud often involves reworking legacy pipelines. The Data Engineering Agent automates this process, converting scripts and proprietary formats into optimized BigQuery workflows. Vodafone, for example, completed a full migration with 100% automation and saw a 90% reduction in migration time, underscoring the agent’s value for modernization projects.
- Unified codebase: Optimizes and standardizes migrated scripts
- Legacy tool migration: Converts proprietary pipelines to native BigQuery workflows
What’s Next: Expanding Capabilities and Integrations
Google plans to enhance the agent with even smarter troubleshooting, integration with development environments, and expanded orchestration through Cloud Composer. The vision is to cover the complete data engineering lifecycle, from development and deployment to monitoring and optimization, making the agent an essential part of the modern data stack.
How to Get Started
The Data Engineering Agent is now available in BigQuery Studio and Dataform UI. Use the ‘Ask Agent’ feature to begin, refer to official documentation for detailed guidance, and share your feedback to influence future enhancements.
- Access the agent: Navigate to BigQuery Pipelines in BigQuery Studio or the Dataform UI. The Data Engineering Agent is accessible via the ‘Ask Agent’ button.
- Learn more: Review the official documentation for setup instructions and best practices.
- Feedback: Email us at [email protected]
Takeaway
The Data Engineering Agent marks a new era for data engineering, where automation and intelligence free teams to drive innovation. For enterprises seeking to accelerate analytics and streamline operations, this agent is poised to be an indispensable ally.
Source: Google Cloud Blog

Google BigQuery’s Data Engineering Agent Is Automating the Future of Analytics