<?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/daa28affbbf94c60ac6a70e01837bc9f&quot; frameborder=&quot;0&quot; width=&quot;1658&quot; height=&quot;1243&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1243</height><width>1658</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1243</thumbnail_height><thumbnail_width>1658</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/daa28affbbf94c60ac6a70e01837bc9f-c24a5df2eba6677b.gif</thumbnail_url><duration>440.32</duration><title>Building a Safe AI Pet Care Planner</title><description>Hi, I am Qiqi, and this is PowPowAI. I started from our homework 2 deterministic healthcare schedule, then added three AI functions: ask me questions for a specific pet via a RAG pipeline with guardrails and self critic scoring, an agent that plans a week for pets using Python tools and returns a structured plan, and an evaluation flow with confidence and critical reports. I demonstrated safe handling of dangerous prompts like ibuprofen in feeding intent, and we logged blocked behavior. I did not ask viewers to take action.</description></oembed>