Author: Jose S
Why data teams use separate dev, test, and production environments
Laptop truth is not production truth. Learn what dev, stage, and prod mean for data work, a simple promotion path from notebook to trusted table, secrets hygiene, and…
Goodhart’s law: what happens when a metric becomes a target
Goodhart’s law: when a measure becomes a target, it stops being a good measure. Pair KPIs with incentives, audit gaming paths, and prefer metrics that are hard to…
How to add notes and callouts that make a chart’s point obvious
Treat titles, callouts, reference lines, a so-what line, and footnotes as part of the chart. Annotation turns a clean picture into a decision aid without lying.
Uncertainty and confidence intervals explained in plain English
Read point estimates with intervals. Confidence intervals show a range of plausible values under a model. Statistical significance is not business importance.
How to set data retention rules so you stop keeping data you don’t need
Treat retention as a keep, archive, or delete decision with owners and dates. Prefer delete, anonymize, or aggregate over forever. Legal holds are a different job from “we…
