Semantic Layer: make your data speak with AI without hallucinations
summary
Describes using a semantic layer to query and present organizational data with AI while preventing hallucinations. Explains combining semantic models, metadata (schemas/ontologies), and retrieval techniques to ground responses in actual data, plus practical approaches to integrate these layers with existing databases and ML tools.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:06BI evolution with AI
- 4:43LLM hallucinations on data
- 9:19What is semantic layer
- 15:18Integrating semantic layer and LLM
- 19:11Demo without semantic layer
- 20:55Demo with semantic layer
- 27:01Agentic loop and advanced demo
- 36:12Best practices and pitfalls
- 40:27Q&A
related talks
17m"Data readiness" is a Myth: Reliable AI with an Agentic Semantic Layer
30mBuild an AI knowledge fabric for your organisation
32mIntelligent Data Platform : Powered by Semantic Layer, MCP & Agents
36mFuture Data Architecture Enabled by New AI Technology
39mLLMs and Hallucinations: Understand, Measure, Control
41mMake It Make Semantic Sense: Semantic Strategies for All of Us
23mTrusted Agentic AI:Architecting Systems with Reliable Data at Scale
35mUnlocking Business Insights with Natural Language and Generative AI