View Resume

Self-Service Tooling · F&G · 2026

Digital Underwriting Guide

  • Scales output

Over 100 pages of underwriting requirements, turned into an interactive guide: an agent enters a few numbers and sees which underwriting class their client could fall into.

All projects

At a glance

Pages of underwriting tables replaced by a few inputs
100+
  • Self-service tooling
  • Interactive guide
  • Agent experience

The challenge

To estimate which underwriting class a client might qualify for, agents had to cross-reference several massive tables spread across more than 100 pages of underwriting requirements — slow work that stood between an agent and a confident conversation with their client.

A spread from the 100+ page underwriting requirements — the dense tables agents had to cross-reference by hand.

The approach

I launched a digital underwriting guide that does the cross-referencing for the agent. Instead of working through the tables, an agent enters a few numbers and the guide shows which underwriting class their client could fall into.

Where AI helped

I used AI as a tool on the repetitive parts of this project. The decisions, and what shipped, stayed with me.

  • Extracting the underwriting tables from a legacy PDF into JSON arrays.
  • Refactoring the first working version into configuration arrays and smaller functions, so the rules live in data instead of logic.
  • Translating components from React to vanilla JavaScript, which cut dependencies.
  • Troubleshooting integrations.

A person reviewed the extracted JSON for accuracy.

How I use AI, and the guardrails around it

The results

  • More than 100 pages of underwriting tables replaced by a few inputs.
  • Agents get an underwriting class estimate without cross-referencing tables by hand.