Why this blog exists

Most of what I learn at work disappears. I solve something hard on a Tuesday, feel like a genius for one glorious afternoon, and a month later I am googling my own problem and clicking a Stack Overflow answer that turns out to be mine, from two years ago, with no explanation. The fix shipped, the ticket closed, and the reasoning that got me there lives nowhere.

This is where I keep the reasoning.

The first reader I write for is me in six months, staring at the same class of problem and wishing past me had bothered to write down why the obvious approach fell apart. Past me never does. Hence the blog.

The second reader is you, assuming you find this stuff as fun as I do. There is a specific kind of person who reads "find the best price and cancel any order, both in constant time" and leans in instead of closing the tab. If that is you, hello. I write these as much to argue with you as to explain to you, so when I get something wrong, tell me. I collect counterexamples the way other people collect mugs.

Everything here comes from code I wrote and ran. The numbers are measured on my own machine, not lifted from a paper or a vendor slide that has never met a cache miss. When I claim a structure is constant time, there is a benchmark behind it, and when the benchmark embarrasses me I leave the embarrassment in. I would rather publish one post I can defend than ten I half-remember.

So the posts are narrow. One data structure, one bug that ate a day, one decision I got confidently wrong on the first attempt. If you came for hot takes on which framework died this week, wrong blog, sorry. If you want to watch a real system misbehave until somebody finally measures it, pull up a chair.