video1.0<iframe src="https://www.loom.com/embed/8d0d0c5253794399ac75ef62a092622b" frameborder="0" width="1548" height="1161" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>11611548Loomhttps://www.loom.com11611548https://cdn.loom.com/sessions/thumbnails/8d0d0c5253794399ac75ef62a092622b-f6ad0c4a3c6f53fb.gif299.947In my Loom I present Paw-Pow Plus, an AI powered pet care planning system. It schedules daily tasks, then uses a RAG augmented agent to resolve conflicts and apply pet specific knowledge with hard safety guarantees baked into Python tool executors. The stack uses Grox and a Llama 3 370B model, plus TF IDF retrieval from a 39 chunk knowledge base for topics like arthritis, anxiety, and medication. The agent has five tools, reschedule, add tasks, root tasks, search knowledge, and mark the result. I include demos in my GitHub for three scenarios, and I share the reliability and guardrail approach. I do not request any action from viewers.