Designing Nophin Commercial Real Estate Underwriter 1st software.
Nophin is a Y Combinator–backed startup focused on helping real estate professionals extract, structure, and analyze property financial documents for faster underwriting decisions.
I worked on product design across the document extraction and underwriting workflow. My focus was reducing manual spreadsheet work, improving financial clarity, and making complex analysis easier to navigate.
Nophin is software used to automate your data entries, analyze data, and streamline underwriting to close deals faster.
And when I mean deals what It means is in the Commercial real estate industry there are not a lot of softwares that investors can used to process Trailing-12 and rent roll documents.
Rent rolls and Trailing-12 are documents commercial real estate owners input the data’s of expenses for 100 to 200 properties from January to December. The documents can be cumbersome and when done properly can be used by real estate investors to sell Assets to investors.
My Role on this project
When I joined the team the entire Idea was in another direction and uncertain.
The interface was looking like google’s homepage. it would have looked like what chatgpt looked like, but chatgpt didn’t exist then. The idea was to design a software that reads data and heavy documents from real estate and have AI spit out operating expenses, rental income. But investors had to tell us thier process and this led me to a path where I had to design something differently.
The Problem
Commercial real estate underwriting often depends on fragmented spreadsheets, rent rolls, T-12 reports, and manual financial interpretation.
Users were spending too much time:
manually entering operating and purchase metrics
switching between uploaded documents and analysis views
validating extracted data before underwriting
translating raw financial tables into usable business decisions
The product challenge was not simply file upload. It was reducing the cognitive burden between raw financial documents and decision-ready underwriting insight.
Creating a home for every deal
A page for all the deals
Product Thinking
The workflow had to support how analysts actually work.
Users do not begin from clean structured data. They usually begin with messy spreadsheets, inconsistent formatting, and multiple source files.
That meant the product needed to support three stages:
Extraction — convert raw uploaded documents into structured financial information
Validation — allow users to inspect, verify, and correct extracted values
Underwriting — transform structured financial data into decision-support outputs
This workflow became the foundation of the product experience.
Underwriting dashboard
What I Learned
A major lesson from this work was that financial tools should not try to hide complexity.
Good financial product design should:
- structure complexity
- reduce ambiguity
- preserve traceability
- support confidence before action
The strongest product decision was treating extraction, validation, and underwriting as one connected workflow rather than separate product surfaces.
Up Next
AirproHVAC is a California-based HVAC business serving residential and commercial clients. The goal of this project was to design a high-converting marketing website that communicates trust quickly, simplifies service discovery, and increases service bookings.
Led the redesign of a cybersecurity learning platform and designed an AI-powered quiz system that transformed passive course consumption into an interactive, feedback-driven learning experience. Worked on a consistent design system for Appsec engineer and redesigned the Webflow website singlehandedly
