Latent Space: The AI Engineer Podcast ⚡️The Rise and Fall of the Vector DB Category
1753 snips
May 1, 2025 
Jo Kristian Bergum, a seasoned search infrastructure expert with two decades at Yahoo and Fast Search & Transfer, dives deep into the evolution of vector databases. He discusses the surge in vector database popularity post-ChatGPT and the misconceptions surrounding embedding-based similarity search. The conversation explores the dynamic interplay between traditional search methods and embedding techniques. Additionally, Joe sheds light on the future of retrieval-augmented generation and the importance of knowledge graphs in AI development.
AI Snips
Chapters
Transcript
Episode notes
Vector DBs Rise from Misconception
- The rise of vector databases was driven by a misconception linking embeddings as the only way to do retrieval for LLMs.
- Vector DBs filled a niche but the category is now converging with traditional search technologies.
Search Technology Converges
- Vector search is now a feature embedded in many traditional databases and search engines, diluting the need for a separate vector DB category.
- The future lies in a convergence where search is the natural abstraction connecting AI with knowledge.
Choose Search System by Scale
- Use PostgreSQL with pgvector for moderate scale vector search if you already use it as your database.
- For critical search business needs, consider specialized search engines for better search quality.