<?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/52c90fd5fb3a45d19b424927d50e60b1&quot; frameborder=&quot;0&quot; width=&quot;1672&quot; height=&quot;1254&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1254</height><width>1672</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1254</thumbnail_height><thumbnail_width>1672</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/52c90fd5fb3a45d19b424927d50e60b1-5f07750a1075c7f0.gif</thumbnail_url><duration>64.5599</duration><title>Building a Clinical Rules Engine with GPT-5 and Codex</title><description>In this video, I share our development of a clinical rules engine designed to deliver the right clinical rules at the right time using GPT-5 and Codex. We structured the rules into an intermediate format and encoded them into Python modules for deployment. Additionally, we built a CDS services compliant API to ensure timely delivery of these rules at the point of prescribing. I encourage you to explore how this integration can enhance our clinical decision-making processes. Your feedback and insights on this initiative would be greatly appreciated.</description></oembed>