Why we built Elofyn

Elofyn started the way most studios start: as a habit that wanted a name. For the better part of a year I shipped small things on Fridays — a script to clean up my inbox, a tracker that noticed when an AI model changed its pricing page overnight, a 900-word essay on a question a friend kept asking me. By the time the habit needed a logo, it had also accidentally become a company.

This post is for anyone wondering why a one-person studio exists in 2026, when so much of the software industry is pointed at the opposite — bigger teams, larger rounds, broader platforms. The honest answer is that I prefer the work that small studios do, and I wanted to make more of it.

What "practical" actually means

Elofyn calls itself a studio that ships practical software, sharp writing, and useful tools. "Practical" is doing a lot of work in that sentence, so it's worth pinning down.

By practical I mean: a thing you can describe in one sentence to a person who is busy. The AI Tool Radar watches a fixed list of AI products and tells you when one of them changes something that matters — pricing, limits, a new model under the hood. That's the whole pitch. The Friday essay on how to read a model card is that essay and nothing else; it does not invite you to sign up for a course. Each release earns its place on its own terms.

The opposite of practical isn't ambitious. The opposite is aspirational — software whose value lives in the imagined future of "and then you could…", in the feature roadmap rather than in the thing you can use today. We try to avoid building anything whose value depends on a sentence that hasn't been written yet.

The cadence is the product

The first design decision Elofyn made was the release cadence: one ship per week, every week. That choice shapes everything downstream.

A weekly cadence means a release has to be small enough to finish, but useful enough to be worth shipping. It means we can't promise a feature that needs three months of work — we can promise the first useful version in a fortnight. It also means that if a week's work doesn't make it, the miss is small. There's another Friday coming.

Most of the studios I admire ship on a regular beat — a record label, a literary quarterly, a weekly newsletter. The beat is the contract with the reader. It says: come back next week, there will be something here. You don't have to wonder if we're still doing this.

A studio that ships every week is a different kind of promise than a studio that ships when the round is closed.

A small set of beliefs

The work assumes these things, and they're worth saying out loud so they can be argued with later:

• Software is most useful when it has an opinion. We'd rather ship a tool that does one specific thing well than a "platform" that lets you build that thing yourself.
• Writing is part of the product. A tool people don't understand is a tool they don't use, and the best explanation usually fits in a long blog post.
• A small audience is allowed. Not everything we make needs to scale. Some tools are worth building for a hundred people who need them exactly.
• The studio is a person. There is no "we" hiding a thousand-person org chart. The byline is the work.

None of these are clever. They're just the constraints that make the weekly cadence possible.

What we're not

A few things Elofyn is explicitly not, so we can stop pretending the question is open:

We are not an agency. We don't take client work — the inbox is open for notes, but not for briefs. We are not a venture-backed startup; the studio is bootstrapped on consulting savings and pays for itself from the first month. We are not optimising for acquisition. If a tool finds the right audience, it will keep existing for as long as it's worth running.

We are also not particularly interested in "AI" as a category, even though one of our products tracks the AI tools industry. Most of our work is plain web software with carefully chosen abstractions. When AI is the right answer, we use it. When a grep and a cron job would do, we use those.

The next release

If any of this is interesting, the best thing to do is come back next Friday. There will be something new on the journal, and a note in the changelog of whichever tool got attention that week. If you'd rather not check manually, the journal has an RSS feed (it lands in the next phase).

That's the whole brief. Practical software, sharp writing, useful tools — shipped every week, in print.