Google embeds AI agents deep into its data stack - here's what they can do for you

Google's new AI agents are not just chatbots; they are autonomous problem-solvers designed for enterprise use. These agents integrate with Google Cloud's BigQuery, Spanner, and Gemini to transform analytics, coding, and infrastructure management.

ZDNET's key takeaways

• Google is introducing powerful technology for AI agents and data management.
• A series of data-centric AI agents are being launched.
• A new command-line AI coding tool is now available.
I am no stranger to hyperbolic claims from tech companies. Anyone who's on the receiving end of a firehose of press announcements related to AI understands. Everything is game-changing, world-changing, the most, the best, yada, yada, yada. And then there's Google. Google is no stranger to hyperbole. But when a company so steeped in data management as part of its core DNA talks about "fundamental transformation," and says that the world is changing because, "It's being re-engineered in real-time by data and AI," we can consider those claims as fairly credible. Also: Got 6 hours? This free AI training from Google can boost your resume today Just in time for Google Cloud Next Tokyo 2025, Google is making a series of announcements that herald a major change in how enterprises manage data. Yasmeen Ahmad, Google's managing director of Data Cloud, says in a blog post, "The way we interact with data is undergoing a fundamental transformation, moving beyond human-led analysis to a collaborative partnership with intelligent agents." She calls this the agentic shift, describing it as "a new era where specialized AI agents work autonomously and cooperatively to unlock insights at a scale and speed that was previously unimaginable." Also: 5 ways to successfully integrate AI agents into your workplace From almost any other company, claims like this would seem like just so much hot air. But Google is backing these claims with real-world capabilities for data scientists and engineers.

The agentic shift

There’s a fine line between AI chatbots and AI agents. Chatbots are conversational, while agents perform autonomous tasks. Some users employ chatbots to perform tasks, as I did when I used ChatGPT to analyze business data. Agents, like ChatGPT Agent, use conversational interfaces to receive instructions. Think of agents as surrogate team members. One agent might handle data normalization (cleaning data), another migration. Each agent performs defined tasks using AI capabilities. Also: Want AI agents to work together? The Linux Foundation has a plan Google envisions agents that automate and simplify tasks for data workers, communicate with each other, and free professionals from tedious work so they can focus on higher-value tasks. Google also aims for agents to work together in virtual teams. There are questions about whether agents free senior professionals or replace junior roles. From my perspective, agents handling grunt work free me to focus on projects and writing.

Cognitive foundation

Traditional databases aren’t enough to feed these agents. Agents need access to both historical and live data, reasoning across silos. Classic data management methods like OLTP (online transaction processing) and OLAP (online analytical processing) isolate data too much for AI to gain insights from trends and current activities. Google has enhanced its database offerings to unify these capabilities. A few years ago, it added a columnar engine for AlloyDB, a fully managed PostgreSQL-compatible database service. Also: How AI agents can generate $450 billion by 2028 - and what stands in the way A columnar engine queries specific columns, reading only needed fields, leading to faster queries and vectorized execution (operations applied to entire columns). Now, Google is adding a columnar engine to Spanner, its globally distributed, strongly consistent database service designed for enterprises needing global reach and high transactional integrity. This also enhances BigQuery, Google’s serverless, scalable, cost-effective multi-cloud data warehouse for fast SQL-like queries on large datasets. Google says this new columnar capability in Spanner speeds up analytical queries by about 200x on live transactional data, enabling instant responsiveness to real-time situations. When building enterprise AI systems, agents must make decisions based on real data. Acting on hallucinated data can cause problems. This is where RAG (retrieval augmented generation) comes in, combining large language models with real-time data access. Also: 5 ways to be a great AI agent manager, according to business leaders Vectorizing search in Spanner and BigQuery is necessary when feeding real-time and historical data. Google adds adaptive filtering in AlloyDB to maintain vector indexes and optimize fast queries on live data. Google introduces autonomous vector embeddings and generation to BigQuery, automatically preparing and indexing multimodal data for vector search—a key step toward semantic memory for agents. BigQuery now supports running AI queries inside the platform, allowing users to ask complex questions (including subjective ones like "Which customers are frustrated?") and get answers directly within analytics tools.

New agentic capabilities

Google announces new capabilities embedding agents into its major data tools:
• Data engineering agent: Simplifies and automates complex data pipelines in BigQuery, driven by natural-language prompts from ingestion to transformation and quality assessment.
• Spanner migration agent: Simplifies migration from legacy systems to BigQuery, automating a normally tedious and risky process.
• Data science agent: Triggers autonomous analytical workflows including exploratory analysis, data cleaning, feature engineering, machine learning predictions, and more. It plans, executes code, reasons about results, and presents findings while allowing user feedback and collaboration.
• Code interpreter: An enhancement of last year’s conversational analytics agent, it converts business-analysis questions into Python code for custom analysis within Google Data Cloud, secured by Google’s infrastructure. An API is available for developers to integrate these capabilities.

New command-line coding tool

Google introduces an extension to Gemini CLI called Gemini CLI GitHub Actions. CLI (command line interface) is a terminal interface used heavily by coders for faster control than graphical menus. Last month, Google made Gemini chatbot features available in the terminal. Now, Gemini CLI GitHub Actions adds agentic features in the terminal environment. Compared to Jules, Google’s coding agent, which works in a secure cloud VM and handles large projects, Gemini CLI GitHub Actions focuses on intelligent issue triage, accelerated pull-request reviews, and on-demand collaboration within GitHub workflows. Issue triage helps manage bug reports and feature requests. Pull requests are GitHub’s method to confirm integrating code changes. On-demand collaboration sets up chat sessions to discuss code. I see programmers using both: Jules for big projects and Gemini CLI GitHub Actions for quick fixes and updates.

Frequently Asked Questions (FAQ)

What are Google's new AI agents designed for?

Google's new AI agents are designed for enterprise use as autonomous problem-solvers, going beyond simple chatbots.

How do these AI agents integrate with Google Cloud?

These agents integrate with Google Cloud services like BigQuery, Spanner, and Gemini to enhance analytics, coding, and infrastructure management.

What is the "agentic shift" that Google refers to?

The agentic shift is described as a new era where specialized AI agents work autonomously and cooperatively to unlock insights at an unprecedented scale and speed.

What is the difference between an AI chatbot and an AI agent?

AI chatbots are conversational, while AI agents perform autonomous tasks. Agents use conversational interfaces to receive instructions but operate independently to complete tasks.

How do the columnar engines in Spanner and BigQuery benefit AI agents?

The columnar engines speed up analytical queries on live transactional data, enabling faster and more responsive AI operations.

What is RAG and why is it important for AI systems?

RAG (retrieval augmented generation) is crucial for enterprise AI systems as it combines large language models with real-time data access to ensure decisions are based on accurate data, preventing problems caused by hallucinations.

What are the specific new agentic capabilities Google has announced?

Google has announced a data engineering agent, a Spanner migration agent, a data science agent, and enhancements to its code interpreter.

What is the purpose of Gemini CLI GitHub Actions?

Gemini CLI GitHub Actions extends Gemini CLI with agentic features, focusing on issue triage, pull-request reviews, and collaboration within GitHub workflows.

Crypto Market AI's Take

Google's advancements in embedding AI agents directly into their data stack represent a significant leap in enterprise data management and analytics. This move aligns with the broader trend of AI-driven automation across various industries, including finance. At Crypto Market AI, we understand the power of intelligent agents to process vast datasets and identify complex patterns, which is directly applicable to the volatile and data-rich cryptocurrency market. Our own development focuses on leveraging AI for sophisticated market analysis, risk management, and automated trading strategies, mirroring Google's commitment to enhancing capabilities through AI. The integration of AI agents into core data infrastructure signifies a future where data analysis and operational tasks become increasingly autonomous, potentially leading to more efficient and sophisticated operations in sectors like cryptocurrency trading and investment. For those looking to understand how AI is reshaping data utilization, exploring our resources on AI-powered crypto trading bots and the broader impact of AI on financial markets can provide valuable context.

More to Read:

• AI Agents: Capabilities, Risks, and the Growing Role in Business Automation
• How to Use Google Gemini for Smarter Crypto Trading
• The AI Gig Economy is Here, and It Pays in Crypto

---

Source: Google embeds AI agents deep into its data stack - here's what they can do for you by David Gewirtz, ZDNET.