video1.0<iframe src="https://www.loom.com/embed/61259d4455ff4db8a9509c3b348e6892" frameborder="0" width="1280" height="960" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>9601280Loomhttps://www.loom.com9601280https://cdn.loom.com/sessions/thumbnails/61259d4455ff4db8a9509c3b348e6892-511a4709280c0a54.gif82.491This Loom explains a project that automatically classifies, enriches, and routes inbound customer support messages using AI. It describes a separation-of-concerns file structure with distinct components for data, prompt design, API call, routing logic, and orchestration. The prompt design specifies a clear role, compiles with allowed value constraints, and uses two few-shot examples for the toughest edge case, with temperature set to zero for consistency. Routing is handled with pure Python for deterministic, testable behavior, and the results.json output contains five fully structured records ready for a downstream team.