Overview
When a disaster wipes out a business, someone has to prove that businesses income loss. DeepWell builds that number and helps defend it in court. Most of the heavy lifting is automated. But AI canโt commit anything on its own, because a named expert has to sign the result and defend every figure in a deposition. So the product runs on automation and still depends on people stepping in at the right moments to check a decision, make a judgment call or sign off.
I came on to this project with the task of redesigning their v1. The first version showed every piece of machinery at once. As the only designer on the team, I redesigned the product end to end and built it in code.
2026 (in active development)
Context & Scope
DeepWell was built by a group of advisors who do this work today, one claim at a time. The engine already existed when I joined, but the experience was lacking, so I led the design from there.
The core challenge was density and trust at the same time. Behind every claim sits a deep stack of documents, rules and AI reasoning that can never be hidden, and the first version showed all of it at once, leaving analysts holding the whole machine in their heads. My job was to let the automation fade into the background while making every human checkpoint clear, deliberate and impossible to miss.
Key Decisions and Tradeoffs
The app opens on what needs a person and lets everything settled recede. The AIโs work shows up as outcomes rather than process, and the heavier a human decision is, the more deliberate the interaction becomes: validating a document gets a second check, changing a judgment call asks why, and approving a report spells out what happens next before it lets you confirm.
Inside a single complex claim, the product turns dense on purpose, and a simple visual rule keeps it readable. Figures from the claimantโs documents are solid, while anything a person or the engine decided is striped, so evidence and judgment read differently at a glance.
The Product
DeepWell is a web app in dark mode grayscale with a bright orange saved for whatever matters most on the screen. Built in React, Vite and Tailwind with Claude Code.
What's Next
v1 of DeepWell only scratched the surface, I am excited to see where we can take this next. And that's all I can say for now!





