Two careers, one throughline.

From environmental engineering…

My background is in environmental engineering — the kind of regulated, document-heavy work where a single project can generate thousands of pages of permits, monitoring reports, sampling data, and correspondence, and where a detail in the wrong place has real consequences. I spent years inside that world, learning how the work actually gets done: the reviews and QA/QC, the paper trail, the regulatory deadlines that don't move, and the standards and regulations that live half in manuals and half in the heads of the people who've been there longest. I cared about that work enough to build software for it — Green Owl Compliance — aimed squarely at the document grind it runs on.

…to production AI

From there I went deep on artificial intelligence — not prompt-engineering demos, but the unglamorous systems work of getting models to behave reliably on messy, real-world documents. I learned where these tools genuinely shine, where they quietly fail, and how much of the job is everything around the model: the data, the guardrails, the evaluation, and the plumbing that decides whether something is trustworthy enough to put in front of a client.

Why Green Owl

Most people selling AI to environmental and engineering firms have only ever seen one side of this. They understand the technology but have never sweated a permit deadline or had a filing bounce back — so they build impressive demos on top of operational chaos, and the pilot dies the moment it meets a real project. I started Green Owl to work the other way around: begin with how your firm actually operates, fix what's brittle underneath, and add AI only where it earns its place.

It's a deliberately small practice. I'd rather do a few engagements well than scale into something that looks like every other AI vendor.