Google Cloud has announced the general availability of Cortex Framework v7, a release aimed at making enterprise data, particularly from SAP systems, ready for AI agents. The framework is designed to address the challenge of turning raw transactional records from ERP systems into semantically rich, AI-ready data products without disrupting mission-critical operations.

Cortex Framework v7 provides purpose-built data product accelerators for SAP that deploy directly into BigQuery and Knowledge Catalog. These data products feed the Gemini Enterprise Agent Platform with accurate business context, enabling AI agents to reason, orchestrate, and execute workflows with high fidelity. The framework translates raw SAP tables into clear business terms, dynamically ingests custom fields, and handles advanced logic such as SAP TCURX currency decimal shifts, maintaining the data fidelity required for LLM interactions and enterprise analytics.

A key addition is the agentic data product builder, which allows data and knowledge engineers to use natural language to generate custom data products. Using the framework’s provided content and Google Antigravity with Gemini, an AI agent can automate the build and deployment process, adapting to specific source data requirements and customizations. The release also supports a clean separation between framework-delivered content and custom models, allowing users to apply updates without impacting custom work.

The modernized data pipelines in v7 are powered by Dataform, offering version-controlled SQL and native dependency management. When a data product is selected for deployment, the framework processes only the required tables, with dependency resolution handling order-of-operations automatically. This modular approach extends to multi-system SAP environments, compiling and deploying data products for SAP ECC, S/4HANA, and SAP BDC in parallel, while dynamic schema discovery ingests custom SAP fields like Z-fields.

Cost efficiency is addressed through BigQuery incremental loading and non-destructive schema updates, processing only new or modified data to minimize compute time and operational expenses. The serverless, BigQuery-native execution scales without additional infrastructure overhead. A bring-your-own change data capture (CDC) approach is also supported, allowing existing CDC-processed pipelines to connect directly.

Following the GA of SAP Business Data Cloud Connect for BigQuery, Cortex Framework v7 natively supports SAP BDC standard and custom data products. Combining these with framework-delivered products can uncover new business opportunities, and v7 includes solution samples across both data types to answer critical questions like sales pipeline health and procurement inefficiencies.

SLB, an existing customer, sees potential in the new capabilities, particularly the agentic data product builder, to accelerate and scale the creation of enterprise data products for analytics and agentic AI use cases. Cortex Framework v7 is available for deployment via demo, and further resources include technical documentation and Google Cloud account team consultations.