Google is adding a Comments to SQL feature to BigQuery Studio in preview, which takes natural language instructions written inside SQL comments and turns them into executable query snippets. The idea is for analysts to formulate intent in «human» language, and Gemini will refine the syntax.
Google is adding a Comments to SQL feature to BigQuery Studio in preview, which takes natural language instructions written inside SQL comments and turns them into executable query snippets. The idea is for analysts to formulate intent in «human» language, and Gemini will refine the syntax.
According to InfoWorld, to get started, you need to enable the SQL Generation widget in Studio and then describe the required columns, sources, and filters directly in the comments between /* … */. After that, the user clicks the Gemini button in the editor and selects Convert comments to SQL. BigQuery shows the generated code and a diff comparison so that you can see exactly how the text prompts were converted to SQL, and you can refine the instructions until the result is correct.
Google explains that this approach should eliminate the routine where the most time-consuming part is «translating» the idea into JOINs, time logic, window functions, and repeating patterns. At the same time, the article emphasizes that this is not a «magic button» for a query like «give me the top products by category for 2023»: the tool is more likely to help with parts of the query and requires clear hints in the comments.
Google has been consistently layering Gemini features into BigQuery, including SQL assist, query explanation, and SQL-native AI functions for working with unstructured data. The trend is not unique: Databricks is developing AI Functions for inference with SQL/Python, Snowflake is promoting Cortex and AISQL, and other storage vendors are also adding «AI add-ons» on top of classic SQL.
Previously, dev.ua wrote about how Google has begun rolling out a new feature that allows users to change their primary @gmail.com address without creating a new account. This is the first time in the service’s 22 years of existence that a user can change their email without losing their emails, files, and settings.
This is a fascinating development. Transforming English comments into SQL queries could greatly simplify data analysis, particularly for people who struggle with traditional coding. I'm curious how this tool can affect accessibility in data science—could it encourage more non-technical users to participate? https://slopefree.io