Skip to main content

Improve ML capabilities with pgvector and Amazon Aurora PostgreSQL

Improve ML capabilities with pgvector and Amazon Aurora PostgreSQL (Level 200)
Majority of the organizational data resides in relational databases, and the need to make this data accessible for training, as well as using ML models to generate predictions in database-based applications has also increased, which may result in more resources and time to support application requirements. In this session, we explain how Amazon Aurora PostgreSQL-compatible Edition and Amazon RDS for PostgreSQL offer support for pgvector, an open-source extension for PostgreSQL, so you can easily store, search, index, and query huge volumes of ML embeddings to power your generative AI applications. Find out how to leverage pgvector to store and search embeddings from Amazon Bedrock, Amazon SageMaker, and more, providing you the ability to build new content, enable hyper-personalization, and create interactive experiences. We demonstrate how to build and deploy an AI-powered application with pgvector and Amazon Aurora PostgreSQL-Compatible Edition for sentiment analysis use case, without the need to build custom integrations, move data around or learn separate tools.
Speaker: Roneel Kumar, Senior Database Solutions Architect, AWS