AIArchitectureEditorEngineering
How I built my AI Site Editor
The architecture, safeguards, and product decisions I used to make AI editing reliable.
Shipping notes, product decisions, and technical experiments from real-world builds.
The architecture, safeguards, and product decisions I used to make AI editing reliable.
Why I started this project, what I got wrong early, and what finally made it useful.
What changed when we moved Gemini Nano from demo mode into real user workflows.
What actually works when you want AI output to feel live inside a visual editor.
Strict JSON output looks clean on paper, but token-by-token delivery turns it fragile in production.
Chat feels magical early, but property panels keep teams precise once complexity grows.