{"slug": "i-built-a-multi-agent-ai-data-analyst-on-google-cloud-that-turns-natural-into", "title": "I Built a Multi-Agent AI Data Analyst on Google Cloud That Turns Natural Language into Business…", "summary": "A developer built a multi-agent AI data analyst on Google Cloud that converts natural language business questions into SQL queries, charts, and strategic reports, using BigQuery's public TheLook E-commerce dataset. The project originated from the Google Cloud Gen AI Academy Cohort 1 Hackathon and integrates multiple cloud services to make database insights accessible to non-technical business teams.", "body_md": "It sounds like a simple business question. Yet in many organizations, answering it still requires a chain of dependencies:\n\nBusiness Team → Data Analyst → SQL Query → Dashboard → Report → Decision.\n\nSomewhere along that chain, valuable time is lost.\n\nI wondered, **What if the database itself became conversational?**\n\nWhat if someone from sales, marketing, or leadership could simply ask questions in plain English and receive not only numbers, but charts, strategic recommendations, and even business-ready reports?\n\nThat question eventually became the foundation of my **Multi-Agent E-commerce Data Orchestrator**, a cloud-native AI system built entirely on Google Cloud.\n\nInterestingly, this project wasn’t born because I wanted to build “another AI application.”\n\nIt began because I wanted to **understand Google Cloud properly.**\n\nEarlier this year, I participated in the **Google Cloud Gen AI Academy — Cohort 1 Hackathon**, organized with Hack2Skill. Rather than treating it as just another competition, I saw it as an opportunity to understand how modern AI applications are actually built.\n\nThe academy wasn’t only about prompting large language models. It introduced the complete cloud ecosystem surrounding AI applications:\n\nAlong the way, I completed learning paths and earned skill badges including:\n\nWhat fascinated me most wasn’t any single technology.\n\nIt was how each Google Cloud service solved a different piece of the engineering puzzle.\n\nInstead of learning them individually, I decided to combine them into one complete system.\n\nThat became this project.\n\nBusinesses today collect enormous amounts of data- orders, customers, products, revenue, inventory, marketing metrics.\n\nIronically, the biggest challenge isn’t collecting data anymore.\n\nIt’s **making it accessible.**\n\nIn the current e-commerce landscape, data is useless if business development and sales teams cannot easily access or understand it. Most business users do not know SQL, and static dashboards often fail when faced with non-predefined questions.\n\nI wanted to remove this bottleneck. I didn’t want just another AI chatbot that summarizes data; I wanted an AI that could “think” through business problems, execute queries, and generate actionable insights.\n\nFor example:\n\n“Find our highest and lowest performing product categories, calculate the revenue gap, recommend whether the lowest performer should be discounted, and save the recommendation as an executive report.”\n\nThat’s no longer a simple database lookup.\n\nIt’s reasoning.\n\nBuilding within the Google Cloud ecosystem allowed me to focus on application logic rather than infrastructure management.\n\nEach service solved a very specific engineering problem.\n\nThat is one of the biggest strengths of modern cloud-native development.\n\nOne of the earliest design decisions involved choosing the data source. I could have created my own database, generated synthetic records, or imported CSV files. But since the project was built on Google Cloud, I wanted the entire stack to stay within the Google ecosystem, that's when I explore one of its most underrated resources — **BigQuery Public Datasets**.\n\nIts a collection of publicly accessible datasets covering domains such as healthcare, finance, transportation, weather, geospatial analytics, machine learning, GitHub activity, Stack Overflow, and e-commerce. These datasets are maintained within BigQuery itself, allowing developers to explore production-scale data without spending time collecting, cleaning, or importing datasets. Storage is hosted by Google, and you only pay for the queries you run (with a generous free usage tier for experimentation).\n\nFor this project, I selected **TheLook E-commerce** dataset, a fictional online clothing retailer developed by the Google Looker team. Although the business itself is fictional, the dataset is intentionally designed to resemble the relational data model of a real e-commerce company. It contains information about customers, products, orders, inventory, logistics, distribution centers, website events, and digital marketing activity, making it an excellent playground for analytics, SQL practice, business intelligence, and AI applications.\n\nUnlike many sample datasets that consist of a single CSV or flat table, TheLook models how production databases are actually structured. Data is distributed across multiple related tables, requiring joins, aggregations, filtering, and relationship mapping to answer business questions, making it much closer to a production database.\n\nThis also meant the AI agent had to understand relationships between tables rather than simply querying one spreadsheet. This made it an ideal benchmark for building a Text-to-SQL multi-agent system capable of answering real business questions.\n\nSide Note:One of the hidden gems of Google Cloud is its extensive collection of BigQuery Public Datasets which are updated periodically. If you’re looking to build cloud or AI portfolio projects, they’re a fantastic starting point. You can explore the available datasets in the official[documentation and browse the]BigQuery Public Datasets[dataset for yourself.]TheLook E-commerce\n\nThe complete **TheLook E-commerce** dataset consists of **seven interconnected tables**. However, for this project, I focused on the three tables required for product and revenue analytics:\n\n```\n                  Products                     ▲                     │ product_id                     │                 Order_Items                     ▲                     │ order_id                     │                   Orders\n```\n\nAlthough the complete dataset is much richer, these three tables were sufficient for demonstrating end-to-end business intelligence workflows such as revenue analysis, category comparisons, and sales recommendations. Like *“Which product category generated the highest revenue?”* or *“Compare revenue between Jeans and Sweaters.”*\n\nInitially, it might seem reasonable to let a single LLM handle everything.\n\nAfter all, modern language models can write SQL. They can summarize results. They can generate reports. So why introduce multiple agents?\n\nBecause different responsibilities require different behavior.\n\nTo achieve reliable performance, I moved away from monolithic workflows and adopted an “Agent-as-a-Tool” architecture.\n\nI wanted one AI to think like a **business manager**, while another focused purely on **database operations.**\n\nThis separation makes the system significantly more reliable.\n\nInstead of mixing planning, SQL generation, visualization, and reporting into one enormous prompt, each agent has a clearly defined responsibility.\n\nThe Manager is the only AI that communicates with the user.\n\nIts responsibilities include:\n\nImportantly —\n\n**The Manager never writes SQL directly.**\n\nThe Specialist exists for one purpose. Working with data.\n\nIts responsibilities include:\n\nIt does **not** perform business reasoning. It simply answers database questions accurately.\n\nEach component should do one job well.\n\nExactly the same philosophy applies to AI agents.\n\n```\nArchitectual               User                            Diagram                    │                            ▼                   Streamlit Interface                            │                            ▼                      Manager Agent                  (Reasoning & Planning)                            │                  Delegates Database Task                            │                            ▼                   Data Specialist Agent                  (Text → SQL Generation)                            │                            ▼                       Google BigQuery                            │                      SQL Query Results                            │                            ▼                      Manager Agent                            │                ┌───────────┬─────────────┐                ▼           ▼             ▼              Business    Charts       Reports              Insights                |───────────|─────────────|                            ▼                       Streamlit UI\n```\n\nRather than relying on one enormous AI prompt, the application becomes an orchestrated workflow.\n\nLet’s look at a typical interaction. If a user asks:\n\n“Find the revenue for Jeans, find the revenue for Sweaters, and compare them in a chart”\n\nThe application doesn’t simply guess.\n\n“Identify our best and worst performing product categories. Calculate the revenue difference and recommend what we should do.”\n\nRather than returning only numbers, the Manager produces a business recommendation explaining whether discounts, bundling strategies, or inventory optimization may improve performance.\n\n“Find our most expensive product and draft a three-sentence cold outreach email targeting VIP customers.”\n\nThe Manager first retrieved the product information, then transformed raw database values into marketing-ready copy.\n\nThat’s where multi-agent reasoning becomes much more interesting than traditional dashboards.\n\nIf you’ve made it this far, thanks for sticking around.\n\nFrom the outside, this looks like a chatbot that talks to a database.\n\nBehind the scenes, it’s prompt engineering, agent orchestration, cloud infrastructure, SQL generation, debugging sessions at unreasonable hours, and a surprising number of redeployments.\n\nThe screenshots don’t show the times the agents confidently generated SQL for columns that didn’t exist, Cloud Run deployments that failed for reasons that seemed completely unrelated, or me wondering why I thought giving AI *multiple* agents would somehow make my life easier.\n\nIt eventually worked.\n\nMostly.\n\nAnd that’s the part I intentionally skipped in this article.\n\nThis was the “how it works” story. The next one is the “how I got it to work” story — covering prompt engineering, Text-to-SQL generation, Cloud Run deployment, IAM permissions, debugging bizarre agent behavior, and all the engineering decisions that transformed a cool prototype into something reliable enough to demonstrate.\n\nSee you in Part 2. Hopefully with fewer hallucinations… from both the AI and me. 😄\n\n[I Built a Multi-Agent AI Data Analyst on Google Cloud That Turns Natural Language into Business…](https://pub.towardsai.net/i-built-a-multi-agent-ai-data-analyst-on-google-cloud-that-turns-natural-language-into-business-368ac87ef6cb) was originally published in [Towards AI](https://pub.towardsai.net) on Medium, where people are continuing the conversation by highlighting and responding to this story.", "url": "https://wpnews.pro/news/i-built-a-multi-agent-ai-data-analyst-on-google-cloud-that-turns-natural-into", "canonical_source": "https://pub.towardsai.net/i-built-a-multi-agent-ai-data-analyst-on-google-cloud-that-turns-natural-language-into-business-368ac87ef6cb?source=rss----98111c9905da---4", "published_at": "2026-07-22 03:49:25+00:00", "updated_at": "2026-07-22 03:59:22.148251+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-agents", "ai-tools", "ai-infrastructure", "natural-language-processing"], "entities": ["Google Cloud", "BigQuery", "TheLook E-commerce", "Google Cloud Gen AI Academy", "Hack2Skill", "Google Looker"], "alternates": {"html": "https://wpnews.pro/news/i-built-a-multi-agent-ai-data-analyst-on-google-cloud-that-turns-natural-into", "markdown": "https://wpnews.pro/news/i-built-a-multi-agent-ai-data-analyst-on-google-cloud-that-turns-natural-into.md", "text": "https://wpnews.pro/news/i-built-a-multi-agent-ai-data-analyst-on-google-cloud-that-turns-natural-into.txt", "jsonld": "https://wpnews.pro/news/i-built-a-multi-agent-ai-data-analyst-on-google-cloud-that-turns-natural-into.jsonld"}}