An agent that makes a good decision is not enough in regulated finance. You need to be able to show your work. This session brings an expert in building agent decision pipelines with traceability and control designed in from the start, covering what auditability actually requires at the architecture level and where teams get caught out when a regulator asks how a decision was made. What this session will cover: - What auditability requires from agent pipeline architecture, not just documentation - How to design decision pipelines that are traceable without slowing agents down - Where control mechanisms need to sit in a multi-agent system - What regulators are starting to ask for from agentic decisioning systems The architecture questions raised here are worth answering well before a regulator asks them for you.