Building at the Speed of Thought
summary
Claims a near‑term model that "writes code better than any human" and shows concrete demos: Gemini 2.5 Pro analyzing an 88,000‑token five‑minute video, extracting timestamps and facts; a web app scaffolded and previewed live (Pokemon marketplace) that actually runs; an agent that turns CSVs into exploratory plots and a housing price model with recursive code execution that fixes errors; robot control from video input; smaller runtimes like Gemini Nano and lower‑cost Flashlight ~7–12¢/M tokens.
AI-written summary · about
chapters (14)
- 0:00AI transformation intro
- 1:27Gemini model family
- 3:13Unified architecture cost
- 4:58AI in robotics
- 6:01Gemini sizes pricing
- 7:06Long context windows
- 8:28AI Studio demo
- 11:20Code execution analysis
- 13:27Multimodal Live interaction
- 14:30Web apps with AI
- 15:55AI in dev tools
- 16:58AI in PR workflows
- 18:04Colab agent demo
- 20:15Call to action
related talks
33mNew Renaissance of Building
18mIntroduction to Gemini APIs and AI Studio
56mBuilding Multimodal Apps with Gemini 2 0 and Gemini SDK
15mBuilding with the Gemini API and AI Studio
19mGemini 2.0: Beyond the Basics
30mUnleashing AI in Your Web Apps
01h 45mAI Engineering with the Google Gemini 2.5 Model Family
01h 55mPrompt to Pipeline: Building with Google's Gen Media Stack