Operations first, then software.
I learned how businesses work from the inside. That changed the kind of systems I want to build.
Before the code
I've spent roughly two decades in operations and management.
- Retail
- Old Navy, H&M
- Trades
- Home inspection, specialty coatings
- Now
- Owner, Flagship Detailing, Northeast Florida
Running my own business means I'm responsible for real customers and outcomes: inquiries that come in while I'm working, schedules that change, follow-ups that can't slip and processes that have to survive constant interruption.
Why AI
I wanted better tools for problems I understood first-hand.
That grew into Alderline Systems, then TextStack, the Android client for SSHHIP and a wider interest in AI agents, engineering workflows, developer tools and governance.
AI agents write most of my code. What I bring is the habit of running systems as accountable operations: someone owns the outcome, a failure needs a recovery path, and impressive output isn't the same as dependable performance.
The record so far
- First serious sessions building with AI coding agents.
- The idea: an AI that answers for the business when I can't.
- Alderline Agent work begins.
- First merged contribution to someone else's project.
- VMGA, PackSift and GCO started.
- Seven pull requests merged into Kun Chen's Firstmate.
- Start building the Android client for Kun Chen's SSHHIP; 28 pull requests merged by mid-November.
- GCO v0.1.0 released; Code Witness and a negative-result experiment published.
- My own SMS agent fails and gets rebuilt; standing rules become automated checks. TextStack is public.
What I'm looking for
Conversations with people building agentic products and infrastructure: founders, technical leaders and operators who value implementation, systems thinking and the messy reality of shipping.
I'm open to selective technical projects, collaborations and operating roles where hands-on AI work meets practical business experience.