A simple static event class that lets any deeply nested code update the main form's progressbar without passing references around.

Testing time-based RxJS5 observables in Angular2 without waiting real seconds, using fakeAsync and tick after jasmine.clock let me down.

How I use TypeScript string literals and keyof to make Immutable.js get calls type safe, so property renames break at compile time instead of runtime.

A reusable mock for testing Ionic presentables like modals, alerts and toasts, so you can trigger onDidDismiss and assert what your page does.

Separate development and production config in Ionic2 using a typed environment module, an OpaqueToken and NODE_ENV to pick values at build time.

A small async/await retry middleware with timeout for NodeJS, replacing the tangled promise-retry code I used to talk to an unreliable database.

Using the Performance API with performance.mark and performance.measure to see how your code really runs on users' devices, not just in the lab.

Ever wanted to give some Firebase users special privileges? Here is how to write a Firebase function that sets custom user claims.

Three ways to manage collections in JavaScript models — exposing the raw array, wrapping it in methods, or extending Array for the best of both.

How I migrated an Ionic app with over 500 components to Angular standalone components using the official schematic, and what needed manual fixes.

Claude 3.7 Sonnet was the first time I caught myself delegating context, not lines. AI coding assistants stopped pretending to be autocomplete.

Three years of AI-assisted development later, vibe coding is real. The thinking is yours, the judgment is yours — the typing was never the point.

AI makes junior developers look productive on week two. If we train them for speed alone, we hollow out the next decade of engineering judgment.

The single-agent model has a ceiling. Claude Code's parallel agents let you split a feature into independent pieces and work like a focused team.

How we merged three Ionic/Angular/Capacitor codebases into a single Nx workspace so a small team could stop maintaining the same code in three places.

I had stocks and ETFs across three platforms and no single place to see them. So I built one, and left it open for anyone else to use.

You cannot prompt AI-generated code into being reliable. What works is what always worked with new hires — guardrails that say no.

I stopped pointing the frontier model at my backlog and started using it to write the Skills and guidelines my cheaper models read.