<?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/70b9689b57204da58b8fef0d23c304fe&quot; frameborder=&quot;0&quot; width=&quot;1920&quot; height=&quot;1440&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1440</height><width>1920</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1440</thumbnail_height><thumbnail_width>1920</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/70b9689b57204da58b8fef0d23c304fe-0614f00a38cb03ab.gif</thumbnail_url><duration>1242.83</duration><title>Introducing the Tracemind Ecosystem for Efficient AI Agent Evaluation</title><description>Hi everyone, I&apos;m excited to share my submission for MCP&apos;s first birthday hackathon, which introduces the Tracemind ecosystem designed to streamline the process of building AI agents. Tracemind consists of four key projects, including a zero-code tracing framework and a Gradio UI for data visualization, all aimed at enhancing the efficiency of agent evaluations. I demonstrated how to create custom test datasets and run evaluations on various models, providing insights into costs and performance metrics. I encourage you to explore the documentation for more details and consider how Tracemind can support your agent evaluation processes. If you find it valuable, please show your support for the project!</description></oembed>