AI Appetite checker
• Lifted quote conversion from 29.7% to 38.9% (+9.3 percentage points) • Designed natural-language risk input, so agents describe a risk in plain language instead of filling a rigid form • Built confidence-tagged recommendations, framed as "likely in this carrier's appetite" rather than a guarantee • Tested and dropped a Kanban-style tracking board after research showed agents didn't need it, keeping the product simpler and faster to ship
• Lead Product Designer • 2025 - present
I ran interviews with ADMs and agents, prototyped alongside the research, and used Granola and NotebookLM to turn call recordings into design decisions.
Overview
Designed an AI appetite checker for the Agentero marketplace that shows independent agents which carriers are likely to write a risk, and how they can place it.
Project brief
Carrier appetite lives in scattered PDFs and tribal knowledge, so agents submit risks that get declined or call their account manager to ask. I designed an experience where an agent describes a risk in plain language and receives a ranked set of carriers, each with a confidence rating and a brief rationale for the judgment. Since a recommendation only helps if the agent can act on it, I grouped results by carrier relationship and routed each card to the right next step.
