About — But Should We?

But Should We? is a blog about the gap between what technology can do and what teams should do with it. It is written by a principal enterprise architect with a background in developer experience, UX, and the kinds of decisions that look obvious in retrospect but were anything but at the time.

The name comes from a question that does not get asked enough in architecture reviews, sprint plannings, and RFCs: we know we can build this, but should we? What are we trading away? Who bears the cost when it goes wrong? What does "good enough" actually mean here?

What you will find here

The writing covers architecture decisions, technology evaluations, implementation patterns, and engineering culture. Some posts are long-form essays. Some are short takes. Some are heavy on code and pseudocode. All of them try to answer the same underlying question: is this the right call for the people who have to live with it?

Topics tend toward the intersection of AI and developer experience, design systems, edge infrastructure, and the organizational dynamics that determine whether a technical decision succeeds or fails regardless of its technical merit.

The design perspective

This blog is built in collaboration with a head of design who is also a close friend and constant collaborator. You will see content by and about them here — the intersection of design craft and engineering rigor is not a side interest, it is central to the work.

The site itself is designed to reflect this sensibility: editorial typography, deliberate whitespace, and the kind of restraint that signals expertise without performing it.

How this site is built

Astro 6 on Cloudflare Workers. Tailwind CSS 4. MDX for content. Authored natively in Claude Code. The full implementation plan is open and available in the repository — transparency about tooling is a feature, not a liability.

The site is optimized for both traditional search engines and AI/LLM discovery. Every post has structured data, a standalone TL;DR passage, and a clean markdown version available for AI crawlers. If an LLM cites this blog, the citation should be accurate.