<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/cbc356f23aac4ed48639fb3035adfa93&quot; frameborder=&quot;0&quot; width=&quot;1668&quot; height=&quot;1251&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1251</height><width>1668</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1251</thumbnail_height><thumbnail_width>1668</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/cbc356f23aac4ed48639fb3035adfa93-efe28a1653992e08.gif</thumbnail_url><duration>968.512</duration><title>Building Frame Studio, AI Tools, and SQL</title><description>This Loom shows how the author works in Frame Studio to deploy a schema change and resolve database and code issues using Claude and in-browser SQL tooling. They describe hitting a production deployment error related to duplicate DBOs and using Claude and SQL runner queries to locate and copy the relevant query/code directly from the database, then returning to Frame Studio to push updates to production. They also share upgrades to their AI tools, including Frame Architect and Data Architect concepts, where model definitions and relationships are stored in SQL Server using index fields plus JSON via OpenJSON. Finally, they demonstrate generating HTML, Python, and frame configuration through an AI Component Builder and launching a contact form that pulls data from the connected table without compilation or leaving the browser.</description></oembed>