Salesforce Exec: Layered Data Architecture Key to Unlocking Enterprise System of Intelligence

The pursuit of an enterprise system of intelligence is driving the development of layered data architecture, according to Tristan Baker, senior director and head of data architecture at Salesforce Inc. Baker spoke with theCUBE's John Furrier at the Neo4j GraphTalk event about how knowledge graphs, ontologies, and layered data architecture are reshaping the way companies approach AI.

The goal is to deliver a conversational interface that returns trustworthy answers in seconds, but this requires more than just a single database. Different questions demand different retrieval structures, and the solution lies in a stack of technologies, including lakehouses, operational databases, customer profile stores, and graph technologies.

Baker frames the solution as a stack, with familiar lakehouses and operational databases at the bottom, followed by a metadata layer that tracks where the truth about a customer lives across dozens of copies. The graph manages the relationships and context that connect those pieces together.

However, the most under-discussed challenge is governance. As metadata moves up to a higher semantic layer, access control can no longer live only inside individual databases. Master data management and governance go hand in hand, and Baker notes that a semantic description of access policy is needed to ensure data protection.