Will Keleher argues that a surprising share of engineering productivity comes from small pieces of knowledge rather than architecture or deep expertise. Not the obvious claim that “everything is made of smaller things” — the sharper one: some nuggets are valuable on their own, with almost no supporting mental infrastructure required. You don’t need to know Python to get use out of python3 -m http.server.

The examples are the argument:

  • ctrl+r history search becomes fuzzy with fzf; atuin replaces shell history with a searchable SQLite database
  • SELECT without FROM, for probing how a database function actually behaves
  • explain analyze in Postgres and MySQL — it runs the query and reports real performance
  • git log -S pattern (the “pickaxe”) to find commits that added or removed a string; git checkout - to jump back to the previous HEAD
  • Globs (**/*.md) cover most find usage; rg over grep/ack/ag
  • Reusing a Node https.Agent for connection keep-alive, which he calls dramatic for latency
  • Bucketing a metric by Math.floor(Math.log10(n)) to see a distribution instead of an average

Then he makes the turn that most tip-list posts skip. At a company, knowledge has the same atomic shape — which data source to debug a given problem with, who knows which area, where the good docs are, when to scale out manually, the rolling-restart command. The bottleneck isn’t that this knowledge is missing. It’s that nobody distributes it.

His proposed practice: post one trick a day to the engineering Slack, technical and company-specific mixed. One per day was deliberately the right cadence — enough to be useful, not enough to overwhelm, and occasionally it started a real discussion. He suggests senior engineers consider doing it.

The best part is the honesty about the format. He concedes you may already know all of these, and that in some domains none of it applies. The claim is about the shape of useful knowledge and the value of circulating it, not about his particular list.