Bring AI based Search to Your Website
summary
Builds a working React app that swaps keyword search for vector search using Weaviate and a few lines of code, ingesting about 10,000 Jeopardy items into a vector DB and showing “kitten” match “cat.” Adds GPT-3.5 for retrieval‑augmented generation to produce hints, translations and feedback, demonstrates multilingual and multimodal image↔text queries, and cites a separate demo at near‑billion vectors to show production scale; the demos run live and the integrations (vector DB + LLM) work as described.
AI-written summary · about
chapters (8)
- 0:00AI search introduction
- 2:28Keyword vs vector demo
- 7:25Semantic search explained
- 13:29Vector databases
- 17:52Multimodal & multilingual search
- 19:42Retrieval augmented generation demo
- 27:37Scaling vector databases
- 29:41Recap and key takeaways
related talks
21mBring AI-Based Search to Your Web App
45mBring Your Own Data! Build Better AI Applications
49mBuilding AI Enabled Web Apps
36mTypescript, You and AI Search
37mBest AI Tools for Web Developers / How to Use AI in Web Development
48mEnter the Brave New World of GenAI with Vector Search
08mRethinking your data stack: The future of AI apps and vector databases
33mEnter the Brave New World of GenAI with Vector Search