How I work
One person does all of this, so method matters more than headcount. Four rules, and they are the reason the work holds up rather than just shipping.
Four rules
I measure instead of guessing
Numbers on this site come from running the thing and reading the result, not from estimating. Where I could not measure something, I say so rather than rounding up.
I check what already exists first
Most problems are already solved by a setting inside software you pay for. Finding that costs you an hour of my time; building a replacement costs you weeks.
I test against reality, not against my own assumptions
A build that compiles is not a build that works, and a passing test written from a guess proves nothing. I check the finished artifact, not the process that produced it.
You own everything
Accounts, data, credentials and code are yours from day one. Nothing I set up depends on me still being around.
What I work in
Tools are means, not the point — but you should know what I can reach for before you ask whether something is possible.
- Languages
- Python · PowerShell · JavaScript · GDScript · C++ · C#
- Platforms
- Windows internals · WMI/CIM · Win32 · web · Godot 4
- Automation
- Make · Zapier · n8n · CRM and calendar APIs · webhooks
- Data & docs
- JSON/CSV pipelines · PDF and document generation · reporting
- Practice
- Automated test suites · mutation testing · profiling · DNS and mail configuration
- Trade
- Carpentry · repairs · installation · fabrication · hauling
What you get, in writing
Every engagement ends in a document you keep, and it is the same shape every time. Five parts:
Findings
What I examined and what I found, stated plainly, including the things that turned out to be working correctly.
What you already own
Capability that is present and unused in software you are already paying for, listed before anything new is proposed.
Recommendation, costed
What to do, in priority order, with the cost and the time attached to each item rather than a single figure at the bottom.
What I could not verify
Every assumption the recommendation rests on that I was unable to test, named explicitly rather than omitted.
The immediate next step
One thing that can be started this week, whether or not you engage me to do it.
Start with a call
Fifteen minutes. If there is nothing worth chasing, I will tell you.
Call or text 419-617-5226zty@zed-tech.net