BigQuery ML lets you build and run machine learning models directly inside BigQuery using SQL. There is no need to move your data out to a separate system or stand up a distinct machine learning pipeline, and there is no requirement to learn a complex framework before you can train a model. For the Professional Cloud Database Engineer exam, the useful thing to register is that this capability exists and what it is used for at a high level, rather than the mechanics of building any particular model.
The core idea is that the model lives where the data already lives. Your data is in BigQuery, and BigQuery ML lets you train and use machine learning models against it without exporting it anywhere. You drive that through SQL, which is why BigQuery ML is described as low-code rather than no-code. You still write SQL statements to train a model and to run it, but you are working in a language you already use to query the data, not writing application code in a separate tool.
That framing matters because it explains who finds BigQuery ML appealing. It tends to suit people who are comfortable with SQL and would rather not write Python when they do not have to, who do not want to move the data, and who do not want to manage separate infrastructure in a different tool just to apply a model. The data stays in place, the workflow stays in SQL, and BigQuery handles the underlying execution.
BigQuery ML is a powerful feature in practice, but the Professional Cloud Database Engineer exam treats it at a high level. The expectation is that you know it exists and can describe what it is for, which is training and running machine learning models with SQL directly inside BigQuery, on data that does not have to leave BigQuery. If a question hinges on recognizing the tool rather than configuring it, that high-level understanding is enough. We would not spend study time going deeper than that for this exam.
Our Professional Cloud Database Engineer course covers BigQuery ML alongside BigQuery as an analytics warehouse and where it fits among Google Cloud's database services, with practice questions that drill these distinctions.
