Build & deploy AI-powered apps
summary
Live demos show how to go from idea to running prototype using AI Studio Build + Antigravity: selected models (Gemini 3.1 Pro/Flash/Flash‑Lite, Gemma 4, VO 3.1 Lite, NanoBanana 2, Lyra 3, Genie 3, OpenClaw) are actually wired up, code exported, and an app scaffolded with Firestore and Google auth and one‑click deploy to Cloud Run — the bookshelf cataloger deployed and persists data. Image analysis runs at well under fraction of a penny per image, and Genie 3 pixel worlds generate navigable 2D frames (not 3D meshes).
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:22DeepMind’s latest AI models
- 6:14AI Studio basics
- 9:02Video analysis with Gemini 3.1 Flashlight
- 13:13AI Studio compare mode
- 16:46URL context for research
- 18:55Gemini 3.1 multimodal
- 19:38Real-time with Gemini Live
- 25:15Build & deploy apps
- 29:37Genie 3 dynamic worlds
- 38:02SVGs from images
- 41:17Image search with Nano Banana 2
- 47:03Video generation with VEO 3.1 Light
- 50:38Gemma 4 open models
- 53:29Music with Lyra 3
- 56:46Bookshelf cataloger demo
- 59:39Gemini Live in robotics
- 1:01:01Q&A
related talks
29mTools and Roadmap for AI-Native Developers
33mNew Renaissance of Building
01h 11mBuild Your Own AI Sidekick with Flutter
37mBuild agents with Gemini API
22mAI-powered JavaScript Development
30mGenAI on Android with Gemini Nano and Firebase
39mBuild the new generation of AI Agents
39mGetting Started with GenAI for Android Apps