Enter The Brave New World of GenAI with Vector Search
summary
Argues that GenAI needs a purpose-built vector store and shows a working pipeline using Cassandra-based Astra Vector DB, embeddings and an embedding service, LangChain-style routing, and approximate nearest neighbor search. Creates a keyspace, a table with a float vector column, a storage-attached index configured for dot product/cosine, inserts sample 5‑dim vectors and runs similarity queries; live demo has glitches but the vector search and retrieval-augmented generation flow executes. Notes GPT hit over a million users in five days.
AI-written summary · about
chapters (19)
- 0:00GenAI and vector search intro
- 3:01AI background and evolution
- 7:43What is generative AI
- 11:37Data scientists vs engineers
- 13:20Understanding GPT
- 17:35NLP and large language models
- 21:27Java frameworks for GenAI
- 23:54Vector databases matter
- 27:43Vector embeddings and similarity
- 31:17Traditional DB limits in AI
- 32:20RAG with vector databases
- 34:03Live demo: Astra Vector Search
- 36:32Create vector DB in Astra
- 41:29Vector columns and custom indexes
- 46:09Insert data and vector search
- 48:39Query by vector similarity
- 51:33GenAI benefits and challenges
- 53:16Get started with Astra Vector
- 54:40Q&A
related talks
33mEnter the Brave New World of GenAI with Vector Search
48mEnter the Brave New World of GenAI with Vector Search
33mEnter the Brave New World of GenAI with Vector Search
34mEnter the Brave New World of GenAI with Vector Search
49mEnter the Brave New World of GenAI with Vector Search
33mVector Databases for GenAI applications
01h 46mVector Databases (for GenAI & beyond)
27mUse AI Vector Search to Build GenAI Apps with Enterprise Data