Beyond Text-to-SQL: Autonomous Research-Driven Database Exploration with DAR

TL;DR

DAR is an autonomous multi-agent system that conducts end-to-end database research within BigQuery, enabling rapid, exploratory data analysis without human prompts, significantly outperforming manual analysis times.

Contribution

This paper introduces DAR, a novel multi-agent system that autonomously performs research-driven database exploration directly inside BigQuery, advancing beyond traditional reactive text-to-SQL systems.

Findings

DAR completes analysis in 16 minutes versus 8.5 hours for humans

DAR produces useful insights and recommendations rapidly

System operates entirely within BigQuery, ensuring data governance

Abstract

Large language models can already query databases, yet most existing systems remain reactive: they rely on explicit user prompts and do not actively explore data. We introduce DAR (Data Agnostic Researcher), a multi-agent system that performs end-to-end database research without human-initiated queries. DAR orchestrates specialized AI agents across three layers: initialization (intent inference and metadata extraction), execution (SQL and AI-based query synthesis with iterative validation), and synthesis (report generation with built-in quality control). All reasoning is executed directly inside BigQuery using native generative AI functions, eliminating data movement and preserving data governance. On a realistic asset-incident dataset, DAR completes the full analytical task in 16 minutes, compared to 8.5 hours for a professional analyst (approximately 32x times faster), while producing useful pattern-based insights and evidence-grounded recommendations. Although human experts continue to offer deeper contextual interpretation, DAR excels at rapid exploratory analysis. Overall, this work shifts database interaction from query-driven assistance toward autonomous, research-driven exploration within cloud data warehouses.