Is Test Blindness Another Second-Order Effect of AI Coding Agents?
Six months into agent adoption, I measured what changed in one busy repo alongside patterns I've been seeing in reviews. Here's what stood out and what comes next.
All of my long-form thoughts on programming, leadership, product design, and more, collected in chronological order.
Six months into agent adoption, I measured what changed in one busy repo alongside patterns I've been seeing in reviews. Here's what stood out and what comes next.
Today I am formalizing TRU - a new metric for Frontend teams that shows how quickly you can produce reviewable work. By definition, it's also a good measurement of how quickly your team can iterate on ideas and bugfixes too.
Simple tweaks to the options in the release profile can dramatically reduce the footprint of the resulting binary
Using workspaces to manage crates and dependencies within a default `napi-rs` project.
Making a feature faster than the node equivalent, whilst adding more capabilities
With no build-step in sight, it can be tempting to put more and more type information into comments.
After a deep dive, I now consider both to be dangerous.
A problem that even the strictest Typescript/ESLint settings cannot help with.
Showing how you can re-use the same name for types and values with Zod. Includes sample in Typescript + JSDoc
Step-by-step explanation of use-cases for the `?` operator in Rust
As a serial stop/start blogger I've written in countless places in the past. The following links represent the ones I stayed around long enough for

Previous articles written on Dev.to
dev.to/shakyshane
Previous articles written on Medium
medium.com/@shakyShane