Google Unveils AI Agent Suite to Automate Data Workflows Google has launched a comprehensive suite of AI agents designed to automate time-consuming data management tasks that traditionally burden engineering teams. The company announced six new "agent" tools capable of autonomously building data pipelines, debugging code, and answering complex business questions without requiring any SQL coding. At the Google Cloud Next Tokyo event, the tech giant introduced these AI-powered solutions aimed at simplifying workflows for data engineers, scientists, analysts and developers across enterprise environments. The company emphasizes that this represents a paradigm shift in enterprise data management - where intelligent agents should proactively meet users where they work, understand their requirements, and handle most of the heavy lifting automatically rather than passively waiting for queries. Starting with BigQuery, Google's new Data Engineering Agent can construct complete data pipelines from simple prompts. Users can provide instructions like "Clean this CSV file, merge it with sales data and push to BigQuery" and the system will execute the entire workflow. For advanced applications, the Data Science Agent in BigQuery Notebooks enables end-to-end model development with visualization and inference capabilities. Analysts and business teams now benefit from enhanced Conversational Analytics Agents featuring a Python code interpreter. This allows users to ask sophisticated questions such as "Segment customers by Q2 behavior" and receive code, charts and insights directly within enterprise security frameworks. Developers also gain new capabilities through Gemini CLI GitHub Actions, an open-source agent integrated into code repositories. This tool can categorize issues, review pull requests and even execute specific tasks through @mentions while maintaining compatibility with security identity joint technologies. A major infrastructure advancement is the AI Query Engine in BigQuery, enabling SQL-based LLM-style queries with hybrid semantic search, vector embedding generation, and a new Spanner columnar engine that bridges OLTP and OLAP capabilities. These tools now handle unstructured data formats including images, audio and video alongside traditional databases. The new Spanner columnar engine accelerates analytical queries by up to 200 times, while BigQuery introduces multimodal tables for combined storage and querying of diverse data types. Google has also opened APIs to enable developers to integrate these conversational analytics capabilities into custom applications, adopting open standards like Model Context Protocol for cross-platform compatibility. Most of these agent features are now available and included in existing BigQuery and Looker pricing models without additional charges.