A business run on software I created, end to end
A mobile tire business had run nearly a year on less than a job a week, with no website and no way to tell which calls were worth paying for. Fixing that meant building the site first, then the measurement behind it. As calls and revenue grew, the bottleneck moved to the paperwork behind every job — six manual systems to reconcile. Nothing off the shelf met the owner's needs, so I wrote the web app the business now runs on daily. Once the inbound work was measured and running, the question left was how much work is out there at all, which meant sizing the market itself.
This wasn't a list of separate engagements. It was one problem followed in order — and each step narrowed what the business actually needed built.
Joining GA4, Google Ads, and Local Services Ads to job outcomes. What made spend defensible — and surfaced where the real constraint was.
Read case study Then · Custom software · Workers, D1, R2Quoting, invoicing, inventory, expenses, and fleet accounts on one job record. Schema design, 25+ migrations against live data, and an integration suite that gates every deploy.
Read case study Now · Market sizing · Python, MapLibreNo dataset counts tire blowouts, so I modeled exposure instead: federal highway data ranked into corridors, plus the registered fleets nearby.
Read case studyWriting the app is half of it. The rest is everything that keeps a live system safe to change.