<?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/18f285422dea46ffaf99bde9c54a95d8&quot; frameborder=&quot;0&quot; width=&quot;1728&quot; height=&quot;1296&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1296</height><width>1728</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1296</thumbnail_height><thumbnail_width>1728</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/18f285422dea46ffaf99bde9c54a95d8-836789ce27ab37e1.gif</thumbnail_url><duration>297.419</duration><title>baole405/kiros - Google Chrome - 9 February 2026</title><description>In this video, I walk you through my submission for the Kairos Technical Assessment, specifically focusing on the AI support triage and recovery hub. I designed a solution to handle slow AI processes, ensuring users never wait by implementing asynchronous worker architectures with real-time updates, achieving API response times under 100 milliseconds. I demonstrate how a ticket is created and processed in real-time, showcasing the use of socket.io for instant updates without manual refreshing. I also explain the backend setup, including the non-blocking API and dedicated workers for AI processing. I invite you to explore this innovative approach and consider its potential applications.</description></oembed>