Hybrid Search & Analytic Processing: Building Databases for the AI Era

Speaker

VeloDB / Apache Doris

Host

CSAIL

Abstract: Modern AI applications demand more than traditional analytics—they require real-time, multi-modal retrieval that unifies structured data, text search, and vector semantics. In this talk, we explore the evolution from customer-facing analytics to agent-facing intelligence, and how Hybrid Search & Analytic Processing (HSAP) enables AI systems to reason and act on enterprise data. Using Apache Doris as a case study, we will dive into core capabilities such as real-time analytics, hybrid retrieval, and lakehouse workloads, and discuss how a unified database architecture can power the next generation of AI agents and data-driven applications.

Bio: Rayner Chen, Apache Doris PMC Chair & VPE@VeloDB, 10 years of experience in distributed system, focusing on distributed scalable analytical databases. Now primarily overseeing Lakehouse-related development.

----

Please reach out to markakis@mit.edu for the Zoom password.