The engineer you hire is the engineer who ships.
Steven Day Tyler, Texas Remote
I have built software for twenty-five years, starting in 2001, and I still write production code every day. I work where product decisions, architecture, delivery and operation belong close together. On multi-year platform engagements, I have been the sole engineer. Two of those engagements have each run for seven years. Mobile, web and operational systems all ship from the same desk.
What I actually build
Three lanes, and they are not separate careers. The same person who writes the offline sync also writes the API it syncs against and answers for it in production.
Mobile in production
One Flutter and Dart codebase serves iOS and Android on Supabase Postgres. It is live in the App Store and Google Play. Riverpod, go_router and freezed structure it. A local Drift database handles offline-first work, conflict resolution and reconciliation. Swift and Kotlin modules use platform channels for video encoding and camera stabilization. I own the model, API, interface, submissions, releases and operation. Sentry watches production.
Web without ceremony
I build with React and Next.js, and use Astro when a smaller surface is the right answer. Sixteen years of WordPress work includes custom themes and plugins written from scratch, plus malware remediation on compromised production sites. I have also converted two client sites from WordPress to Next.js and deployed the replacements.
Operations need software
I built a field platform for a multi-state emergency response company. Native iOS and Android apps handle dispatch, GPS, scene capture and offline incident reporting. It replaced paper forms and a Microsoft Access database, and has remained in the App Store since 2020. Another platform runs on iPads at three heavy-equipment sites in two states, covering parts, purchasing, rentals, transport and service.
I do not manage people. I set the standard and own what ships.
I have never managed people, and I do not want to. My job is to make the technical calls, turn them into working software and stay accountable after release. I have been product owner and project manager for a six-person outsourced build, writing and prioritising the Jira backlog, running sprint reporting and holding vendor syncs. Before that, I spent twelve years running web and digital operations inside a regional academic health system, working directly with the VP of Information Technology and the Director of Marketing.
I know how to coordinate work. I want my hands on the product.
AI architecture covers how the agent fleet, the gates and the observability actually fit together.
How I work is the method itself, and Work is what came of it, including the engagement that did not go to plan.
Insights holds the long-form papers if you want the full version rather than the summary.
The standard runs before the code ships.
I have codified roughly two hundred architecture standards as machine-checked quality gates. A static analyzer I wrote is published as a public package. Every agent run has session-level observability, and about forty command-line harnesses let an agent operate the real services the work depends on. Two published technical papers document the system. One command is the whole definition of done. Quality is checked by machinery rather than remembered in a review.
If the work needs an owner, I should talk.
I am open to staff-level mobile work, senior full-stack roles, contract engagements and fractional product engineering. The title matters less than the shape of the responsibility. If you need someone who can make the calls, write the code and remain answerable for production, send me the role, the problem and what already exists.