Typescript, You and AI Search
summary
Shows how a Typescript Next.js app uses OpenAI embeddings and a GA Typescript weate client to turn a song list into a semantic search index, storing vector embeddings in a we8 vector DB and returning humanlike results (songs about “summer” or “color”). Demonstrates type-safe collection generics, a multimodal CLIP/Gemini demo, and a 10-line RAG pattern that asks an LLM for a favorite from retrieved results—demos run live and work.
AI-written summary · about
chapters (10)
- 0:00TypeScript AI search intro
- 2:09Keyword search limits
- 7:46Semantic search with ML
- 15:23Weaviate vector database
- 16:27Semantic search workflow
- 19:19Live demo: TypeScript vectors
- 25:31Type safety in Weaviate
- 29:03LLM-augmented search
- 32:23Summary and resources
- 32:45Q&A on AI search
related talks
45mBring Your Own Data! Build Better AI Applications
31mBring AI based Search to Your Website
30mBuilding AI-Powered Apps with NodeJS and Gemini | Armen Vardanyan #JSConfAM25
37mBest AI Tools for Web Developers / How to Use AI in Web Development
40m4 pillars for building a modern AI application in JavaScript
34mJavaScript-Powered AI: Crafting Intelligent Systems with LLMs and Vector Databases
01h 16mAzure SQL ❤️ Open AI
23mWebAI: Building the next user experience in the browser