<?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/4829724605da4bcda13c8a702e47a28a&quot; frameborder=&quot;0&quot; width=&quot;1792&quot; height=&quot;1344&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1344</height><width>1792</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1344</thumbnail_height><thumbnail_width>1792</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/4829724605da4bcda13c8a702e47a28a-5c22d501239f138a.gif</thumbnail_url><duration>281.1974</duration><title>Optimizing AI Rules in Missive</title><description>In this video, I discuss how to efficiently create rules in Missive using the ChatGPT API for message analysis. It&apos;s crucial to ensure that we only analyze messages once to save on costs, as each analysis incurs a fee. I share my approach to crafting a single analysis prompt that can be reused across multiple rules, which helps in minimizing token usage. I also emphasize the importance of categorizing AI rules for better management and consistency. Please take a moment to review these strategies and consider implementing them in your workflow.</description></oembed>