Google Announces New AI Agents for Data Analysis & Science

Google has introduced a suite of six specialized AI agents designed to automate and simplify data analysis and science tasks that traditionally consume significant time and effort from data teams. These new tools leverage advanced AI to build data pipelines, debug code, and answer complex business questions in plain English, eliminating the need for manual SQL coding.

Key Highlights

• Google launches AI agents integrated with BigQuery, GitHub, and Spanner to automate data workflows.
• Gemini-powered APIs enable developers to embed conversational analytics directly into applications.
• BigQuery now supports AI-native SQL and vector search for enhanced query capabilities.
At the recent Google Cloud Next Tokyo event, the company showcased these AI agents aimed at data engineers, scientists, analysts, and developers. This initiative represents Google's vision of an "agentic enterprise," where intelligent agents proactively assist users by understanding their needs and performing heavy lifting tasks.

BigQuery Data Engineering and Science Agents

The Data Engineering Agent in BigQuery can create entire data pipelines from simple prompts such as "Clean this CSV, join it with sales data, and push it into BigQuery." For more advanced needs, the Data Science Agent within BigQuery Notebooks facilitates end-to-end model building, including visualizations and reasoning.

Conversational Analytics Agent

Upgraded for analysts and business teams, this agent now features a Code Interpreter that runs Python code behind the scenes. Users can ask complex questions like "Segment customers by behavior in Q2," and receive the corresponding code, charts, and insights—all within a secure enterprise environment.

Gemini CLI GitHub Actions

Developers benefit from the Gemini CLI GitHub Actions, an open-source AI agent integrated into repositories. It can triage issues, review pull requests, and respond to @mentions to handle specific tasks, seamlessly integrating with Google Cloud’s secure Workload Identity Federation.

Infrastructure Enhancements

BigQuery now includes an AI Query Engine that supports LLM-style queries within SQL. It also offers hybrid semantic search, vector embedding generation, and a new columnar engine for Spanner that bridges OLTP and OLAP workloads. Google is extending these capabilities beyond structured data to unstructured formats such as images, audio, and video. The Spanner columnar engine accelerates analytical queries up to 200 times faster, while BigQuery supports multimodal tables that combine various data types. Additionally, Google is releasing APIs for developers to embed conversational analytics into their own applications and is adopting open standards like the Model Context Protocol to ensure interoperability with other AI tools. Most of these features are rolling out now and are available at no additional cost within existing BigQuery and Looker pricing plans.

---

Originally published at Maginative on August 5, 2025.

---

Frequently Asked Questions (FAQ)

Q: What are the main categories of Google's new AI agents for data analysis? A: Google has introduced agents focused on Data Engineering, Data Science, Conversational Analytics, and developers (Gemini CLI GitHub Actions). Q: How do the BigQuery Data Engineering and Data Science Agents work? A: The Data Engineering Agent can create data pipelines from simple text prompts. The Data Science Agent within BigQuery Notebooks assists with end-to-end model building, including visualizations and reasoning. Q: What can the Conversational Analytics Agent do? A: This agent allows users to ask complex business questions in plain English, and it can generate corresponding code, charts, and insights, leveraging a Code Interpreter that runs Python code. Q: How do developers benefit from Gemini CLI GitHub Actions? A: Developers can integrate these AI agents into their repositories to triage issues, review pull requests, and respond to @mentions, automating specific development tasks. Q: What infrastructure enhancements are included with these AI agents? A: BigQuery now features an AI Query Engine supporting LLM-style SQL queries, hybrid semantic search, and vector embedding generation. Spanner also receives a new columnar engine to accelerate OLTP and OLAP workloads. Q: Are these new AI agent capabilities available at an additional cost? A: No, most of these features are rolling out now and are available at no additional cost within existing BigQuery and Looker pricing plans.

Crypto Market AI's Take

Google's foray into specialized AI agents for data analysis signifies a broader trend in the industry: the democratization of complex data tasks through intelligent automation. This aligns with our mission at Crypto Market AI to leverage AI for enhanced financial decision-making. While Google focuses on enterprise data workflows, our platform, Crypto Market AI, applies similar AI principles to the volatile world of cryptocurrency. Our AI-powered trading bots and AI analysts are designed to navigate market complexities, offering insights and executing strategies with speed and precision, mirroring the efficiency gains Google aims to achieve in data science.

More to Read:

• How to Use Google Gemini for Smarter Crypto Trading
• AI-Driven Crypto Trading Tools Reshape Market Strategies in 2025
• The Future of AI in Finance: Trends and Opportunities