Level: Intermediate to Advanced
Prerequisite: See below
Enterprise data architectures must evolve beyond traditional warehouses to support generative AI applications that require semantic context and unified access patterns for effective retrieval-augmented generation (RAG). Most organizations struggle with extending existing data investments to support GenAI, GenBI, and semantic search applications while maintaining performance and governance.
This comprehensive half-day course provides technical implementation guidance for extending enterprise data architectures through a "context engineering" approach that integrates vector databases, knowledge graphs, and universal semantic layers. Participants learn techniques for understanding production-ready RAG systems, deploying intelligent caching strategies, and building graph-RAG architectures that enhance AI accuracy through business context.
The course covers understanding vector database technologies and optimization for enterprise-scale, hybrid search architectures that combine dense vectors with metadata filtering, knowledge graph implementation for semantic context, and comprehensive operational frameworks for monitoring RAG system performance.
Additionally, an introduction to YAML will provide an overview of defining semantic layers and understanding how to extend existing data architectures to create a unified semantic infrastructure that makes enterprise data truly AI-ready.
You Will Learn:
- Understand context engineering through the RAG approach for improving LLM quality by providing relevant business context and reducing hallucinations in generative AI applications
- Establish comprehensive RAG operations frameworks, including performance monitoring, cost optimization, and quality assurance for extending data architectures to support production AI applications
- Understand vector database technologies and hybrid search architectures for extending existing data platforms with enterprise-scale RAG capabilities for GenAI and semantic search applications
- Implement universal semantic layers using YAML configuration with centralized metrics stores and intelligent caching that extend traditional BI architectures to support GenBI and AI applications
- Understand knowledge graphs and Graph-RAG systems that enhance AI accuracy by connecting existing enterprise data through semantic relationships and business context
Geared To:
- Enterprise data architects
- Data platform engineers
- Technical leaders
- Solution architects
- CDOs and data leaders
- BI architects
Prerequisites
Experience with enterprise data architecture and a basic understanding of semantic modeling concepts. Familiarity with vector databases or similar search concepts is helpful but not required.