Featured
How I decide whether a process is worth automating with AI
The biggest cost isn’t the model. It’s checking what the model produces. Here’s the simple calculation I start every automation conversation with.
Articles
Articles and notes
I don’t write often, and only about what I’ve tested in practice. Each piece answers one specific question.
When Docker Swarm still makes sense, and when it starts costing too much
Swarm isn’t fashionable, but in a small team it often beats Kubernetes. Here’s how I weigh the choice and how I can tell it’s time to change it.
An automated WCAG scan is the start of an audit, not the end
A green result from an automated tool only means it didn’t find what it knows how to find. Here’s what the rest of the audit looks like.
Build your own tool or buy one off the shelf? How I run the numbers
Comparing a subscription to the cost of a single sprint is the most common mistake in “build or buy” decisions. Here’s what else you need to count.
What developers should test, and what QA should test
The “who tests this?” argument usually comes from nobody splitting the responsibility. Here’s a simple split that works well in small teams.
Why I started writing about technology decisions
There are plenty of tutorials. What’s missing are pieces about why someone chose one option over another, and how it turned out a year later.
Nothing published on this topic yet. See all articles.