<?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/1c99c12d7a1f429b8c358b4a449edcf0&quot; frameborder=&quot;0&quot; width=&quot;2218&quot; height=&quot;1663&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1663</height><width>2218</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1663</thumbnail_height><thumbnail_width>2218</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/1c99c12d7a1f429b8c358b4a449edcf0-1caa9f0835c55967.gif</thumbnail_url><duration>312.483</duration><title>GitHub search with natural language</title><description>In this video, I provide an overview of how vectors, embeddings, and track completions work in Chalk, specifically using GitHub data. I demonstrate how to create feature classes and compute various features, including user data and project information. I also show how to generate embeddings based on search queries and retrieve relevant project data. Please take a look and let me know if you have any questions or need further clarification!</description></oembed>