<?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/653510d080f34c75a4c810771f1d9630&quot; frameborder=&quot;0&quot; width=&quot;1324&quot; height=&quot;993&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>993</height><width>1324</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>993</thumbnail_height><thumbnail_width>1324</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/653510d080f34c75a4c810771f1d9630-9d0a4a9df3536933.gif</thumbnail_url><duration>665.949</duration><title>Deploy an OpenAI Agent on Modal</title><description>This Loom shows how to deploy the e-commerce customer support agent built with the OpenAI Agents SDK on Modal. It covers logging into Modal for the initial $30 in free credits, syncing the project in a code space, running modal setup to complete authentication, and creating an OpenAI secret in Modal. The speaker then tests the query script locally, deploys the agent using uv run modal deploy, and verifies the agent response and OpenAI traces showing use of the search policies tool. They close by encouraging viewers to extend the agent with additional tools, better search and retrieval, conversation memory, a chat front end, and other reliability, security, and scheduling considerations.</description></oembed>