Zero-Latency Analytics in Your Application with Dives

BI tools were never built to be app interfaces – they're rigid, clunky, and add complexity to your user experience. Dives are a different way to build data experiences. Dives are interactive data apps you create with natural language — and now you can embed them directly into your own applications. In this livestream, we'll show you how to build a Dive and embed it in your app, with help from your favorite agents, of course. You'll learn how to:   - Create interactive visualizations with natural language queries   - Embed Dives into your app with a secure sandbox   - Set up the auth flow so your users get read-only access without exposing credentials   - Choose between server-side and dual execution with DuckDB-Wasm 00:00 Intro & welcome 00:50 What we'll cover today 02:15 What is a Mother Duck Dive? 04:00 Core use cases: BI, data apps, embedded analytics 06:30 Dives vs traditional embedded BI 09:00 Dual execution architecture explained 11:30 Customer spotlight: Ahead Computing 12:45 Live demo: building a dive with Claude Code 15:30 Under the hood: local preview & dive guide 17:00 Q&A: Is there a semantic layer in Motherduck? 19:45 Q&A: Data modeling & the comment feature 21:30 Embedding a dive on Vercel 25:00 Q&A: Data prep & dimensional models for dives 27:30 Q&A: Versioning dives as code on GitHub 29:15 Embedded dive walkthrough on Vercel 31:00 Interactive map dive (NYC 311 data) 33:30 Customize-your-own dive with in-app AI agent 37:00 Deep dive: dual execution & DuckDB WASM 39:30 Full pivot table dive with Malloy semantic layer 43:30 Earthquake globe: pushing dives to the limit 46:00 Q&A: Will dives be in open source DuckDB? 47:30 Q&A: Malloy models & custom libraries 49:15 Q&A: Multi-tenant setups with separate databases 51:00 Q&A: D3.js support & Excel export 53:00 Q&A: Granular per-user access control 54:30 The dive gallery & community contest 56:30 Wrap up #datavisualization #businessintelligence #duckdb