Rahul Auradkar of Salesforce grew up in India, loving cricket. He devoured as much data as he could about the game from the newspapers, the radio and TV. His voracious appetite for information gave him the context to follow his favorite batsmen and bowlers. Following the game better — isn’t that what it’s about these days? Getting the data, making sure it can be used and applied in a way to “follow the game,” so to speak? Today, Auradkar, the guest on this episode of The New Stack Makers, is executive vice president and general manager at Salesforce. He leads the Data Cloud and Einstein group, also known as the Unified Data Services team. Now, decades later, he realizes that all the data he looked at as a young fan provided him with some context. Today, it’s a lot different. It’s still a mental game, but the tools we use tell us quite a bit about the evolution of structured applications, “old AI” and why data exhaust has become a way to define the application itself. Today, it’s about normalized schemas — to make sense of the data in some normalized way. How Engineers Approach Data Management Normalized schemas speak to the evolution of how engineers have treated data over the past 20 to 30 years. It started with the acceptance and scale-out of enterprise applications. These applications all had relational databases that stored data. Questions then surfaced about how to access that data and how to use it. Enter the data warehouse. Enterprise managers started to question how to obtain and use the data. Engineers built data warehouses to access that data. “Warehouses were analyzing structured data on which you drove analytics,” Auradkar told the Makers audience. “The analytics, over time, started to get fed back into business processes.” But then, what about the data generated from the data warehouses? That data started to get treated as an exhaust that teams would analyze to improve business processes. But then the “aha” moment arrived when the business realized they could take snapshots of the data and apply new concepts such as machine learning, or the “old AI,” as Auradkar called it. The old AI, the boring or predictive AI, changed the way we think about data exhaust. The applications don’t need to treat the data they create as exhaust. That exhaust is how we define the application. “Along the way, what was born?” Auradkar asked. “What was born was databases, warehouses, data lakes, lake houses, all of these were born along the way. Lake houses were born because you wanted to get the best of warehouses and data lakes, which really took unstructured data into account. “Now, all of those ended up having their own storage protocols, their own table, constructs their own file, constructs their own query protocols, even SQL as a language. SQL as a language has its own SQL variant, whether it is done by Oracle or by Microsoft, right? So that’s how the debt got created. “The silos got created, and these silos have a lot of trapped data in there. And how do you untrap it seamlessly so it can be used for meaningful interactions for businesses in their business applications? That problem remains unsolved, and that’s what we are looking to solve.” Please check out the full episode to learn from Auradkar how Salesforce is thinking through its agentic AI platform and Data Cloud.