How vector databases power AI search

Priscilla Parodi

Principal, Community Advocacy

Elastic

Josh Devins

Senior Principal Engineer, Search and ML

Elastic

Search is evolving rapidly. Users expect search boxes to understand the meaning behind search queries. Users ask questions, use everyday language, or maybe even upload pictures to search for products. AI-enabled search with large language models takes what is possible to a whole new level.

A vector database is a key technology that supports nearly everything we might think of as modern search. In this webinar, we’ll look at text search, vector search, and hybrid retrieval, and how vector databases power these use cases.

This webinar is for anyone who wants to build a modern search experience and wants to know what tools to use. The content may be useful for developers, product managers, solution architects, ML technologists, and professionals in related roles.

Highlights

• Separate the buzz from facts: See how search has evolved, and what you need to know about vector databases and vector search
• Look at the use of Elasticsearch as full vector database
• Understand the capabilities you need to build a generative AI-based search experience

Additional resources

• Learn more about vector search and Elasticsearch as a vector database
• Find sample apps and notebooks in the Elasticsearch Labs repo on GitHub
• Create ChatGPT for your private data, using Elasticsearch for vector storage
• Compare BM25 and our sparse encoder retrieval in this relevance workbench demo