AI Systems Builder & Automation Specialist
I design and direct AI-built systems, then verify every piece myself. I make the architecture and the technical calls. AI writes the syntax. I prove it works before it ships. Anyone can prompt a model. Far fewer can prompt it and then check whether the result is actually right.
How I actually use AI: directed to do the reading and drafting, never trusted with a decision that has to be right, and verified before anything ships. The method behind every project here, with the proof pulled from three of them.
View case study →A six-domain production service I architected and shipped on my own. Headless rendering, deterministic scoring, regression-tested, and built to fail loud instead of silent.
View case study →A staged job-search engine. It pulls real postings, scores each one against a candidate with deterministic code, and tailors materials by reordering real bullets. It never invents a qualification.
View case study →Automated pipelines that pull a business's scattered marketing and sales data into one place every day. Idempotent, fail-loud, and built to run unattended for months.
View case study →A research-grounded content engine: pulls real studies, verifies every citation two ways, and won't publish a claim it can't trace to a real paper. Shipped and live.
View case study →The instinct under a deadline is to automate everything. With Role Reaper, the right move was the opposite. I stopped short of full automation on purpose.
Knowing where not to automate is the judgment, not a limitation.
Auto-submitting applications gets them filtered out before a human ever sees them, and the platforms ban it anyway. So the system does everything up to the submit. It finds the fit, tailors the resume, fills the form, and then hands control back to me for the one step where a person in the loop actually wins.
That's the pattern in everything I build. AI writes the syntax, but the decisions are mine. What to build, what not to build, and how to prove it works. A system isn't done when it runs. It's done when I've checked that it does the right thing.
I'm an AI systems builder. I design the system, direct the AI that writes it, and verify every piece before it ships. My background is in psychology, and that's the lens I bring to systems. They exist to change what a person can decide, and they're only done when they actually work, not when they just run.
I run E11EVEN Analytics, and I take on build projects directly as well as full-time roles. I built everything on this site while working a full-time job, which should tell you what I do with my own time.