<?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/ee0455604e524656a37ee641e77d19c2&quot; frameborder=&quot;0&quot; width=&quot;1670&quot; height=&quot;1252&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1252</height><width>1670</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1252</thumbnail_height><thumbnail_width>1670</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/ee0455604e524656a37ee641e77d19c2-b181794b419a61ad.gif</thumbnail_url><duration>80.59</duration><title>Optimizing AI Agent Performance and Troubleshooting Challenges 🚀</title><description>In this video, I discuss the challenges I&apos;ve faced in software engineering over the past 10 years, particularly in optimizing workflows with the introduction of AI agents. I highlight issues like retry storms and contact thrash, which can complicate performance. To address these problems, I&apos;ve utilized tools like Llama Index and Redis for indexing, as well as Horizon 3 for security scans. I emphasize the importance of understanding these issues to improve our systems. I encourage viewers to consider how we can better monitor and enhance our AI agents&apos; performance.</description></oembed>