Building AI Agents with PostgreSQL & pgvector: Vector Search Inside Your Database
Date: Saturday, 26 December 2026
Time: 11:00 AM – 1:00 PM IST
Speaker: Sonu
Technology: Data & AI
Certification Track: AI-900
Discover how to build AI-powered applications and intelligent agents using PostgreSQL and pgvector. This session will introduce participants to vector search and explain how databases can store and retrieve information based on semantic similarity, enabling AI applications to deliver more relevant and context-aware results.
Participants will learn the fundamentals of PostgreSQL, vector embeddings, and the pgvector extension. The session will explore how text and other information can be converted into vector embeddings, stored directly in a PostgreSQL database, and searched based on similarity rather than relying only on traditional keyword-based queries.
The session will also cover database schema design, vector storage, indexing techniques, and similarity search. Through practical demonstrations, participants will understand how to connect vector search with an AI agent to retrieve relevant information and use it to generate more informed responses. This approach is useful for building applications such as AI assistants, knowledge search systems, and retrieval-augmented generation (RAG) workflows.
In addition, the session will discuss performance optimization, security considerations, and production-readiness challenges when implementing vector search within a database. Participants will gain insights into designing efficient retrieval workflows and understanding how PostgreSQL and pgvector can support modern AI application development.
What You Will Learn:
PostgreSQL, vector embeddings, and pgvector fundamentals.
How to design database schemas and store vector embeddings.
Vector indexing techniques and similarity search.
How to build a retrieval workflow for an AI agent.
How vector search can support context-aware AI applications and RAG workflows.
Performance optimization, security, and production considerations.
Practical demonstrations followed by an interactive Q&A.
Join us to explore how traditional databases can power modern AI applications, learn the fundamentals of vector search, and understand how to build AI agents that retrieve relevant information directly from PostgreSQL. The session will conclude with a live demonstration and Q&A.