Daily Digest
Why TDD and AI coding tools exploit the same psychological flaws
Drew DeVault draws a provocative parallel between Test-Driven Development cults and GenAI adoption, arguing both exploit developers' psychological need to feel competent while potentially undermining actual code quality.
- TDD's hidden influence on architecture: While ensuring test coverage, TDD shapes codebases to be "testable" rather than well-designed, and provides no guarantee that passing tests verify the right behavior for actual user needs.
- The dopamine trap of development metrics: Both TDD's green test suites and AI's rapid output create addictive feedback loops through coverage percentages, CI badges, and productivity metrics that make developers feel competent regardless of actual code quality.
- AI coding agents as the new performance theater: GenAI tools let mediocre programmers experience the rush of 10x developer productivity, building "cathedrals" quickly that have beautiful test coverage but rotten foundations underneath.
- The psychological cost of shortcuts: Developers chase these tools despite knowing the externalities (environmental costs, job displacement) because the feeling of finally being "great" at programming overrides long-term concerns.
Latest Articles - Page 5
What is Codemode
More than a year ago I wrote a few posts here that recommended people not to load custom tools into their context (or MCP servers) but to just use more scripts. Most importantly I wrote that Code Is
Fragments: October 4
In response to my last fragments (probably the bit about us worrying if LLMs have consciousness when we when we should be wondering why they don’t have a conscience) “Metalanguage” replied: we shipped
Benchmark In Milliseconds
Benchmark In Milliseconds Oct 5, 2026 How long should a micro benchmark run? My rule of thumb is to tweak the input size until the benchmark takes about 300ms, for the following reasons: 1.31s with 2
A new, bespoke static site generator to replace Jekyll
My blog began as a blosxom (Perl) site running on a VPS. In 2011 I moved to the new GitHub Pages, with the site generated statically by Jekyll (Ruby) on a GitHub server. It was a no-brainer: easier, f
Strands Decider: Why Not an Encoder?
Strands Decider: Why Not an Encoder? Expertise level: exhausted. As I admitted in my last post, I am very much not an expert model developer. What follows is likely to be, at least partially, inaccu
We're going to need default hard budget caps on pretty much everything
No summary available.