Beyond Vector Search: Building Ultra-Resilient GenAI Applications...
summary
Argues that standalone vector stores create operational friction around data synchronization, ACID guarantees and enterprise resilience and demonstrates a PostgreSQL-compatible distributed database using pgvector to implement RAG pipelines that scale multi-region horizontal scaling with ACID semantics. Presents concrete numbers — 249 AI companies in 2024 and exponential MAU growth — and describes ultra-resilience patterns for surviving peak traffic, grey failures and region outages. Describes an implemented system that embeds vector indexes beside application data, supports bursting, and preserves consistency under load.
AI-written summary · about
chapters (10)
- 0:00GenAI and search apps
- 1:06AI use cases growth
- 3:51Designing for exponential change
- 5:36Faster iteration cycles
- 7:43Postgres for GenAI apps
- 9:07GenAI app pipelines
- 10:10Integrating RAG
- 12:36MCP servers access
- 13:41Ultra-resilient GenAI apps
- 15:06Conclusion
related talks
36mUnleashing the Power of Vector Databases for AI-Driven Applications
42mChoosing the Vector Database for Your GenAI Application
57mEnter The Brave New World of GenAI with Vector Search
41mPostgreSQL for AI. Real Vector Search in PostgreSQL
47mExtendable by Design: Building Generative AI Apps with Postgres and Vector Storage and Azure AI
13mRAG Without the Hassle: Building AI-Powered Applications
06mWhat makes vector databases essential for RAG and AI apps?
04mSupercharging GenAI PostgreSQL Vector Search Optimization