video1.0<iframe src="https://www.loom.com/embed/ccc6835178c74de8aecf45466239624c" frameborder="0" width="2904" height="2178" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>21782904Loomhttps://www.loom.com21782904https://cdn.loom.com/sessions/thumbnails/ccc6835178c74de8aecf45466239624c-dc08d6e3548ff725.gif238.02Hi, I am Kyle McDonald, and I walk through how investment teams can get faster, more reliable answers from their data without relying on SQL, dashboards, or waiting on analysts. Behind the scenes, ThoughtSpot queries a governed semantic layer, so definitions like corporate bond spread and volume are consistent, and the first answer took about three seconds. I show joining multiple data sources like spreads and 10 year treasury yields in one governed query, then detecting risk signals like issuers with about 50 basis points widening in the last 30 days. When I rephrase the question, the result stays deterministic and consistent. I also describe a spotter agent idea that would continuously scan for signals like CDS spreads over 300 basis points in the last 72 hours, running on existing data in about three minutes.