video1.0<iframe src="https://www.loom.com/embed/849be6d8995d426bae587c88f1841b50" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/849be6d8995d426bae587c88f1841b50-129d88eed368375d.gif1170.527This Loom demonstrates an enterprise risk management AML workflow, focusing on how the scenario builder drives end-to-end analyst triage. It shows an alert queue with live-model data for Dragon Petrovich, then builds a conditional triage scenario where repeat subthreshold deposit bursts and multi-destination transfers trigger AI agents and logic gates (AND, OR, XOR), escalating to a Fraud Operations Team or requiring secondary lead review. The presenter covers agent autonomy settings (auto route, suggest, require review), scenario run and backtesting, including a test result of 50 records with 25 matches and an estimated 68% false positive rate. It then walks through creating an ATM shell game draft scenario, escalation to lead or secondary review actions, and concludes with a dashboard showing seven high severity cases and seven SLA breaches plus workload and detection health metrics.