Google Unveils Conversational AI Agents and Unified API for Enterprise Data Analytics in BigQuery

Google Cloud has integrated advanced Conversational Analytics and a dedicated "Agents Hub" into BigQuery, enabling enterprise teams to query large-scale datasets using multi-turn, context-aware natural language. The update introduces unified API endpoints and an Agent Development Kit (ADK) designed to streamline the deployment of governed, custom AI agents across corporate applications.

By: AXL Media

Published: Feb 15, 2026, 12:51 PM EST

Source: Information for this report was sourced from InfoWorld https://www.infoworld.com/article/4124972/google-expands-bigquery-with-conversational-agent-and-custom-agent-tools.html

Advancing from Static Queries to Conversational Intelligence

The primary development in Google’s latest BigQuery update is the shift from one-off text-to-SQL commands to a persistent conversational model. The new Conversational Analytics agent, now available in preview within the BigQuery "Agents Hub," allows users to engage in ongoing dialogues with their data. Unlike previous iterations that treated every prompt as an isolated event, this system retains context, including active filters, time ranges, and specific datasets, across a single session. This evolution allows business analysts to refine complex queries progressively through natural language, significantly reducing the technical barrier for non-SQL users.

Strategic Rationale and the Unified API Framework

The launch of the Conversational Analytics API and the Agents Hub addresses a growing strategic need for centralized data governance in the era of generative AI. By providing unified API endpoints, Google is enabling developers to embed these analytics agents directly into third-party applications and internal workflows. This architectural shift minimizes "logic duplication," where different teams might otherwise build disparate systems to interpret user intent. Instead, the unified API ensures that access controls, metric definitions, and security policies established in BigQuery are consistently applied, regardless of where the AI agent is being accessed.

Integrated Tools for Custom Agent Development

To support professional developers, Google has introduced the Agent Development Kit (ADK), a suite of tools that integrates with the Model Context Protocol (MCP). This allows custom-built agents to not only query BigQuery but also connect to a wider ecosystem of over 30 external data sources. Key features within the ADK include "verified queries", previously known as golden queries, which allow data scientists to ground an agent's reasoning in pre-approved, high-accuracy SQL logic. This ensures that when a user asks a high-level business question, the agent prioritizes "trusted" paths to calculate critical metrics like revenue or churn.