Takeaways From Databricks Data + AI Summit And Investor Briefing

研报英文原文证据摘录

Takeaways From Databricks Data + AI Summit And Investor Briefing

erates entirely within Unity

Catalog governance, eliminating the need for additional data movement or synchronization

pipelines.

■LTAP (Lake Transactional/Analytical Processing): LTAP is a new data architecture that

unifies transactional, analytical, and streaming workloads on a single copy of data within

the lakehouse. It combines the Databricks Lakehouse with Lakebase (serverless Postgres) to

support both operational and analytical use cases. The architecture eliminates the need for

ETL pipelines, replicas, and change data capture by enabling direct access to operational

data for analytics. LTAP supports independent scaling of transactional and analytical

workloads while maintaining a single governed data foundation.

■CustomerLake: CustomerLake is an agentic CDP built natively on Databricks to unify

customer data, identity, AI models, and activation capabilities. It enables continuous,

real-time decisioning where agents dynamically analyze and act on customer behavior.

The platform integrates data ingestion, identity resolution, audience segmentation, and

campaign execution within a single governed system. CustomerLake is designed to operate

directly on the lakehouse, allowing models and data pipelines to drive customer engagement

workflows in real time.

■OpenSharing: OpenSharing is an open, vendor-neutral protocol for sharing data and

AI assets across organizations and platforms. It extends the Delta Sharing standard to

include AI models, agent skills, and unstructured data, with support for cross-cloud and on-

premises environments. The protocol provides standardized APIs for discovery, authorization,

and access, enabling organizations to collaborate and exchange assets without custom

integrations.