Hi, I’m Daryll. I build WordPress systems that begin with the why.
Readable architecture, thoughtful processes and DevEx tools that make engineering feel calmer and more predictable.
I’m a staff engineer working on large, long-lived WordPress and editorial systems.
I care about turning business problems into clear plans, shaping architecture that’s safe to change, and helping engineers work with more confidence and less friction in their day-to-day work.
Most of my work sits at the intersection of architecture, developer experience and AI-assisted engineering. I’m especially interested in how structure and clarity make AI more useful, and how small improvements in DevEx can completely change how a team feels.
Controlled Acceleration
AI makes execution cheap. The hard part is making sure fast execution still produces software you understand and trust.
Controlled Acceleration is the model I use for AI-assisted engineering: humans own intent, AI handles execution, and constraints keep everything aligned.
Latest writing
The most recent things I have published. Mostly deep dives. Occasionally, notes and experiments.
-
The 90/10 Problem
AI built most of a recent feature, and built it well. Then UAT surfaced a requirement from a client meeting that never made it into the planning docs. This is about the last 10%: the context that never reached the AI, and the judgement calls about risk and timeline that it can’t make on your…
-
The Model in Practice
Six issues describing five phases, and I don’t use them that way any more. Plan, specify and constrain have collapsed into a single conversation. The collapse is the success condition rather than a failure. The five phases remain the fastest way to diagnose work that has gone sideways, even when they’re no longer sequential steps.
-
Skills as a Type System for AI
Ad-hoc prompts produce inconsistent documentation because the AI doesn’t know your standards. You can hand a junior engineer a style guide; with prompts there is no equivalent. So I built one. Encoding standards as reusable skills works like a type signature — constrain the input and the output, and the shape of the work stops…
How I think about engineering
I have spent most of my career working on WordPress at scale. Multisites. Legacy codebases that need to evolve. Editorial teams with sharp workflows. My focus is on the foundations that let those systems change safely.
I am interested in how AI fits into that work in a real, practical way, not as a toy, but as a teammate that handles the repetitive glue while humans shape the architecture and constraints.
Some principles I try to work by
- Clarity beats cleverness. If I cannot explain it, we should not ship it.
- Structure reduces ambiguity. Typed inputs and outputs make AI and humans both more reliable.
- Developer experience multiplies output. Friction is a tax on every project.
- Calm systems win. Good architecture makes change boring instead of risky.
- AI works best with guardrails. Use it where patterns are clear, and feedback loops are tight.
Things I have built or maintained
A few of the tools and libraries I have worked on over the years. Some started as client needs. Most exist to make engineers lives a little easier.
- Safe SVG – A WordPress plugin that lets you safely upload SVG files to the media library by sanitising inline SVG content. View on wordpress.org →
- SVG sanitiser library – A PHP library that strips unsafe elements and attributes from SVGs so that teams can rely on vector assets without worrying about security surprises. View on GitHub →
- wp-scaffold – A starting point for modern WordPress projects, with structure, tooling, and patterns that aim to reduce friction for teams working across multiple codebases. View on GitHub →
- DevEx and CLI tooling experiments – Internal tools for commands, scaffolding, and AI assisted code generation that try to bake good defaults into the everyday developer workflow.