Future Data Architecture Enabled by New AI Technology
summary
Claims that AI plus vector search will cut dependence on data specialists and automate data catalogs, SQL generation and answers from warehouse-backed knowledge; shows a working SaaS called Unionall that builds data product descriptions, dictionaries and example SQL and publishes storefronts to Snowflake Google and Amazon. Uses embeddings, ChromaDB style vector stores, LLMs and a RAG with domain agents decision layer, converts numbers to text blobs to avoid vector weaknesses and grounds answers to avoid hallucination.
AI-written summary · about
chapters (13)
- 0:00AI-driven data architecture
- 1:52Data architecture challenges
- 7:59Cloud warehouses limitations
- 12:53AI revolution in data interaction
- 16:21Why dashboards fail users
- 20:16RAG architecture explained
- 23:44Numeric data in AI systems
- 25:11Multi-agent AI for domains
- 28:25Better AI responses with prompts
- 30:12Automating data catalogs
- 32:14AI-powered data sharing demo
- 34:02AI impact on data operations
- 34:43Adopting AI in data architecture
related talks
56mDeveloping AI Applications with SQL AI: Ground to Cloud to Fabric
46mBusiness: AI as an enabler for unlocking Industry Data: Monetizing Enterprise Value
10mThe game-changing future for Data and AI
58mAI-Enhanced Big Data: Integrating Private LLMs and Vector Databases
15mMapping the Future of Data & AI
15mUnlock the Power of Your Data for Intelligent AI Applications
01h 00mEnterprise Architecture for Next Gen Innovations
24mFuture of Data and AI