The business introduced major upgrades to its data analytics platform at Google Cloud Next 2025 with the goal of making data work more contextual, conversational, and intelligent.

In an exclusive interview, Google Cloud's VP of Engineering, Peter Bailis, described the company's approach to these advancements and talked about what it refers to as an autonomous data to AI platform.  Google introduced personalized agents for various client roles within BigQuery and Looker. Consider conversational analytics agents in Looker that enable you to query data in plain English, data science agents that support model development in Colab notebooks, and data engineering agents that aid in pipeline construction and maintenance.

These are not only lauded chatbots. Through a new BigQuery Knowledge Engine that comprehends schema connections, business terminology, and query history, they are contextually grounded, trained on your organization's data, and fully integrated into your workflows.

This term "autonomous" refers to more than simply set-it-and-forget-it infrastructure. It implies that you won't have to spend time juggling CSV files or looking for documentation because the system can recommend joins, transformations, and pertinent datasets. Among thousands of users, their BigQuery SQL and Python code helper has a 60% approval rate, which is twice as high as Bailis says is normal for coding models. Over the last nine months, use has increased by 350%.

The timing of Google's offer to test autonomous data on AI platforms couldn't be more ideal for businesses who are currently having trouble with their data foundation. Better technologies are only one aspect of the future of data work; another is a fundamental shift in who can engage in data discussions and how quickly those talks are translated into action.