Webinar RecordingData Agents in BigQuery: From Data to Answers in Natural Language

In this hands-on webinar, we will create a custom data agent in BigQuery that understands your company’s data structure, business terminology, and calculation rules, and transforms natural-language questions into ready-to-use analysis..

100+
happy students
1.5+
video hours
∞
access to the materials

Home
/
Webinars
/
Data-agents in BigQuery: From Data to Answers in Natural Language

programs

Webinar Syllabus

PART 1. THEORY
What we will talk about
How an agent transforms natural-language questions into data analysis and SQL queries.
How to provide an agent with business context, terminology, calculation rules, and validated queries.
The capabilities, limitations, access requirements, security considerations, and cost of using data agents.
The business scenarios in which a data agent can deliver real value.
What You Will Achieve
You will understand how data agents interact with databases and learn how they differ from a standard chat interface connected to a table.
PART 2. PRACTICE
What we will talk about
An introduction to the BigQuery Studio interface and a review of the settings of a ready-made demo agent.
Access management: the roles and permissions required to use the tool.
The simplest way to interact with data: launching a basic chat directly from a data table.
What You Will Achieve
You will become familiar with the Conversational Analytics interface through ready-made solutions and learn which technical permissions are required to get started. You will also understand how the simplest format of quick chat-based queries to tables works, allowing you to retrieve initial data instantly.
What we will talk about
The key difference between a basic chat and a custom AI agent.
A step-by-step process for creating your own data agent from scratch.
How to create detailed instructions, business terminology glossaries, and validated parameterized queries.
How to use BigQuery ML’s built-in artificial intelligence capabilities for forecasting and anomaly detection.
The limitations of the technology and why tables containing raw data must never be connected.
What You Will Achieve
You will learn how to configure an AI agent from scratch based on your project’s business logic. You will also understand how built-in forecasting and anomaly detection algorithms work and how to prepare the data architecture correctly to avoid errors in the AI system’s outputs.