Vector Databases (for GenAI & beyond)
summary
Walks through vector DB fundamentals, shows demos on Astra DB and openAI/huggingface embeddings, and claims practical rules that saved CPU and bugs: use dot on normalized vectors for speed, don’t mix embedding models or thresholds because similarities shift, and plan for approximate search (ANN) since exact KNN fails past ~20 dimensions. Builds demo apps (philosophy quote RAG, typo-tolerant name lookup) and shows migrations and metric mismatches so readers can judge fit.
AI-written summary · about
chapters (30)
- 0:00Introduction
- 1:27Speaker context
- 2:14What is a vector
- 5:27Traditional vs vector queries
- 7:15Vector search examples
- 9:45Vector databases purpose
- 12:16Vector math foundations
- 17:23How vector search works
- 18:07Apache Cassandra overview
- 21:46Similarity metrics: Euclidean vs cosine
- 26:47Exact search challenges
- 29:34Demo 1: Vector experiments
- 34:54Lessons from demo 1
- 37:05Dot product similarity
- 42:50Vector DB performance tips
- 46:01Demo 2: Creative vector search
- 51:28When vectors work
- 52:10Dot product peculiarities
- 56:49Other similarity metrics
- 58:19Demo 3: Migrating vector DBs
- 1:02:34Vector DBs for GenAI RAG
- 1:06:08Vector embeddings
- 1:10:00Demo 4: Text embeddings
- 1:14:39Embedding model comparisons
- 1:19:18Building RAG with vector search
- 1:24:24Advanced vector optimizations
- 1:27:57How vector DBs work
- 1:31:33Measuring performance and accuracy
- 1:34:50JVector engine
- 1:37:00Closing & Q&A
related talks
45mA gentle introduction to Vector Databases
12mVector Databases: Why QA professionals Needs to Care About them in the Age of AI?
50mA fun and absurd introduction to Vector Databases
57mEnter The Brave New World of GenAI with Vector Search
33mEnter the Brave New World of GenAI with Vector Search
01h 23mVector Databases Demystified: Discover how they work
48mA fun and absurd introduction to Vector Databases
48mA Fun & Absurd Introduction to Vector Databases