Course Outline
Foundations of the following concepts:
- Vector mathematics
- AI-generated vector embeddings
- Leading AI embedding models
- The mechanics of semantic search
- Methods for calculating distance metrics
Analysis of vector indexing strategies:
- Implementation of IVFFlat indexes
- Application of HNSW indexes
Deep dive into the PgVector extension for PostgreSQL:
- Setup and installation procedures
- Managing and retrieving high-dimensional vector data
- Evaluation of distance metrics
- Leveraging vector indexes for performance optimization
Learning Objectives: Upon completion, participants will possess a comprehensive understanding of key AI-enhanced PostgreSQL extensions. They will acquire practical expertise in integrating Large Language Models (LLMs) and vector search capabilities into production-grade applications.
Requirements
Foundational proficiency in SQL and prior exposure to PostgreSQL are required.
Training Environment: DaDesktops instances running Linux virtual machines (facilitated by NobleProg).
Intended Audience: Developers specializing in database applications, system architects, and data analysts.
Testimonials (2)
Tuning strategies.
Jeffrey Zieg - Matrix Consulting
Course - PostgreSQL Performance Tuning
Logging behaviour when the instance is under stress, and the hierarchy/nomenclature of instances, databases, files, etc.
