Courses
Code-first courses on building AI that actually ships. Each course is a set of ordered lessons you can read straight through or dip into — and every lesson comes with a runnable, framework-free demo you can open in the browser.
Chrome's Built-in AI — from 0 to superhero
On-device Gemini Nano, handed to your page as plain web APIs: no backend, no API
keys, no per-call cost, and nothing leaving the machine. Seventeen lessons take
you from a first availability() check all the way to shipping — the Prompt
API, structured output and tools, multimodal image input, the task APIs
(Summarizer, Translator, Language Detector, Writer, Rewriter, Proofreader),
on-device embeddings, and the agentic layer (WebMCP, Generative UI, an in-browser
MCP client) — then the production tail: observability and evaluation.
Every lesson has a runnable demo in the companion repo (github.com/danduh/chrome-ai-course), with hosted versions at windowai.danduh.me.