Dart and Flutter and AI, Oh My!
summary
Alpha software that ships a reusable chat widget and provider API called the Flutter AI toolkit, released to GitHub and pub in alpha a few weeks ago, that plugs into Google Gemini provider or Firebase Vertex LLMs, streams responses, accepts multimodal uploads, and runs locally against Gemma/Lama too. Demo shows hot reload integration into a recipe app, JSON-first responses parsed into app models, and simple RAG embeddings search (dot product similarity) to pull grandma’s recipes — it works end to end.
AI-written summary · about
chapters (15)
- 0:00Welcome
- 1:48Speaker credentials
- 3:14Generative AI fundamentals
- 6:44Flutter AI toolkit
- 8:52Live demo setup
- 10:59Adding LLM chat
- 14:53Customizing LLM responses
- 17:50Parsing LLM outputs
- 23:21Adding recipes via chat
- 24:03Retrieval augmented generation
- 28:24Extending chat with RAG
- 35:10Simplifying AI features
- 37:18Magic button with JSON
- 39:06Local LLMs support
- 40:57Resources and next steps
related talks
42mMore Flutter and Dart and AI, Oh My!
36mEmpowering Your Flutter Apps with AI
01h 11mBuild Your Own AI Sidekick with Flutter
14mPrompt to production: Building an AI app with Flutter
38mEnhance the customer experience
47mAgentive Experiences With Flutter: What, How, and Why Bother?
50mFlutter: The Last UI Framework
01h 19mFull Stack AI Features