Complete your Databricks User Groups profile!

Fill out a few details about yourself so the community can get to know you.

Community BrickTalk | Vector Search in the Lakehouse with NEAREST BY

BrickTalks

Oct 29, 8:00 – 9:00 AM (PDT)

89 RSVPs

About this event

Join us for a free, community-sponsored Bricktalk as we dive into how SQL-Native Vector Search works directly within Databricks SQL and Runtime.

While vector search originally developed as a millisecond serving problem for chatbots and search bars, many enterprise workloads, like daily payment transactions or overnight historical record enrichment, are batch-oriented and require similarity searches across millions of records.

In this session, we will explore how Lakehouse vector search brings these workloads directly into Delta and Iceberg tables using NEAREST BY joins in SQL, eliminating the need for separate serving endpoints or dedicated vector databases.

What We’ll Cover:

  • Batch-Oriented Workloads: How vector search applies to enterprise tasks like matching millions of daily transactions or historical data enrichment.

  • SQL Integration: Using NEAREST BY joins to find closest matches across tables directly in Databricks SQL and Runtime.

  • Efficiency & Cost: How optimized execution engines and vector indexes narrow search candidates to boost throughput and lower costs.

Date & Time: Thursday, October 29, 2026

  • PT: 8:00 AM

  • ET: 11:00 PM

  • London (BST/GMT): 3:00 PM

  • IST: 8:30 PM

Register to save your spot

Presentation(s)

Speaker

  • Alex Jiang

    Databricks

    Associate Product Manager

Organizers

  • Mandy Ross

    Databricks

  • Rishabh Tiwari

    Databricks

    Community Engagement Specialist

  • Tushar Parekar

    Databricks

    Program Manager

  • Sujitha Ramamoorthy

    Databricks

    Lead Organizer

Contact Us