About

I've been writing backend and systems code since 2007. Most of that time went into search and indexing, performance work, and modernizing legacy systems without breaking production, where the real constraint is usually the behavior you inherit, not the code you write.

I think in mechanisms. Before I touch anything, I want to understand the constraints, the feedback loops, and the state the system settles into. Simplification comes after that understanding, never instead of it. I'm not afraid of complex systems when the complexity buys something specific. I have no patience for complexity that buys nothing.

Outside of code, the approach is the same; only the material changes. I build my own things: structures and hardware, not just software. The same questions apply: what are the loads, what fails first, what is irreversible, what can be measured. One engineering practice, two toolsets, not a job and a hobby.

I write plainly, skip the preamble, and I'd rather send you a diagram than a paragraph.

DECISION FILTER
dumb
robust
elegant

BUT FIRST
is this worth doing? is the requirement even real?
FIG. 2 — ONE PRACTICE, TWO MATERIALS
software
structures & hardware
loads · failure modes · tolerances
since2007
domainsearch · indexing · perf
stackphp · prestashop · mysql/mariadb · elasticsearch · kafka · redis