Work
The path here wasn't linear, but it made sense in hindsight.
It started at Rakuten — as a data analyst using tools that were clearly built by people who had never had to use them. That friction became curiosity, and curiosity became code. What started as scripts to automate my own work grew into building internal tools the whole team used, then leading a small team, then doing it full time as an engineer. By the time I made the formal switch, I'd already been doing it for a year.
Since 2020 I've been at Autodesk, working on TradeTapp — a risk management platform inside Autodesk's Construction Cloud that helps general contractors evaluate subcontractors before hiring them. Risk scores, financial ratios, safety records, automated questionnaires — the kind of due diligence that determines whether a project succeeds or goes sideways. It's a hard domain with real stakes, and I've spent the last four years as principal engineer building and scaling it.
Right now I'm thinking a lot about AI — not as a feature to bolt on, but as something that changes how we write software and what we can offer customers. Autodesk has spent years acquiring products across the construction lifecycle, and the interesting work now is in the connective tissue: what happens when all that data talks to each other, and how AI changes what's possible at that intersection.
What I keep coming back to: tools that solve real problems for people who have no patience for tools that don't work. The domain almost doesn't matter. The discipline does.
Principal Software Engineer
TradeTapp lives at the intersection of construction, finance, and risk — and it's genuinely complicated. General contractors need to know if a subcontractor is financially stable, has a clean safety record, and has the capacity to take on more work before they sign a contract worth millions. I've been the principal engineer on the team building the platform that makes that possible, sitting inside Autodesk's BuildingConnected network of over a million construction companies. The product is cloud-based, integrated across Autodesk Construction Cloud, and used by some of the largest GCs in the country.
Production Analyst → Software Engineer
This is where I learned that the best way to improve bad tools is to build better ones yourself. I started as an analyst, got frustrated by the workflow, and ended up building the internal tooling that replaced it — a web application for configuring and scheduling client reports, ETLs connecting our data systems, APIs for the machine learning team, and an internal admin portal that saved everyone from drowning in spreadsheets. I wore a lot of hats here and it was exactly the kind of environment where that was possible.