Three content engineering hats in late 2026

Content design is getting the design engineer treatment

Three content engineering hats in late 2026
Photo by Jan Kopřiva / Unsplash

What does a design engineer look like for content designers? I started asking myself that after reading Nurkhon Akhmedov’s What does a design engineer do?.

Nurkhon describes a design engineer on his team who builds prototypes directly in the codebase, tests them with users in 3 to 4 days, and hands engineers a finished component with a Storybook story. He cites a UX Tools survey of 1,478 designers: among people with the title design engineer, 80.9% spend more than half their building time on AI-generated code.

For content designers, the answer is spread across several job titles. A LinkedIn search in November 2026 returned no exact matches for titles like “UX Writing Tools Engineer” or “Content Design Systems Engineer”.

But the work itself does show up: content designers asked to write prompts and evaluation rubrics, small teams running AI content workflows and engineers who own documentation pipelines.

This post groups that work into 3 hats, with job postings and open-source projects for each. Just keep in mind that the postings may have closed by the time you read this.


What “content engineer” means elsewhere

First, a note on the name. If you search for “content engineer”, you’ll mostly find a marketing role: someone who builds AI pipelines that produce content for search and answer engines. Render’s Senior Content Engineer is one example: “the primary audience for this content is coding agents and LLMs”.

Ryan Law at Ahrefs has argued against hiring for that role, but there’s some overlap with the work in this post. Ahrefs’ own explainer on content engineering describes a “structured content engineer” who “designs taxonomies and metadata schemas”, which is close to the structural work further down.

The difference I see is in the goal. Marketing content engineering is measured in traffic and citations, while the hats in this post are about the copy people read inside a product.


What’s changing

In Nurkhon’s model, the design engineer hands off something that already works. The details that often get lost between design and development get fixed by the person who designed them. Nurkhon borrows Karl Koch’s term for them: “polish”, the final 10%.

For content designers, the equivalent would be handing off rules and structures along with the strings.

I already see the work shifting among fellow content designers and in their teams. Writers design and vibe-code, designers and developers write, machine learning engineers push frontend features.

UX writers still write UI strings, though I’d say we do it less and less. More of our guidance now ends up in agent skills, reference files, evaluation rubrics for AI-written text and automatic tests that flag problems in the copy before it's delivered.

I’ve written more about this in What content designers need to know about typed parts and AI content design for UX writers: a day in the life.

What content designers need to know about typed parts
Writing rules for copy we don't control: my first days in the engineers' schema
LLM content design for UX writers: a day in the life
How a UX writer and ML engineer collaborate to generate dynamic, user-facing error messages.

So which hats are content designers, and the people around them, starting to wear? Here are the 3 I found.


Build writing tools for AI

This is the closest match to a design engineer: building tools that apply UX writing standards while engineers and designers work.

An early example is ux-writing-skill, an open-source agent skill built by Christopher Greer, a Staff Content Designer at Stripe. It helps Claude, Codex and Cursor write and edit microcopy against a set of standards, and it’s aimed partly at design system teams that want to “enforce content guidelines at scale without becoming a bottleneck”.

The UX Writing Hub ran a workshop on the same idea in May 2026. It taught content designers to build “guidelines that enforce themselves” with Claude Code skills, such as a tone checker.

I’ve tried small versions of this myself: turning a UX proposal into a Claude skill and testing instruction files for an AI assistant.

Repurpose your UX content assets with AI
The internal doc you built for one approval meeting can have a second life, and you get to own the process.
SOUL.md and AGENTS.md for content designers
What I learned writing and testing agent instructions

Teams were checking copy against rules before AI tools arrived, though. Vale is an open-source tool that “brings code-like linting to prose”: a team turns its style guide into rules, and Vale flags text that breaks them. Elastic checks its docs with Vale every time changes are committed to a pull request.

For product UI copy, Ditto keeps strings in one place, synced with Figma, and its “magic edit” feature checks selected text against the team’s style guide and suggests fixes.

In job postings, this hat rarely gets a title of its own. It shows up inside broader content design roles:


Model the content structure

This work is closer to content strategy and design systems. It means building the structures and pipelines that let the same content feed a website, an app, emails, the help center and a chatbot.

In practice, that includes designing taxonomies and metadata schemas, automating the steps from draft to publication or adding content and brand voice checks to the pipeline. The tools are often low-code, such as n8n, CMS data models and JSON Schema.

The idea has a longer history in technical documentation. Back in 2024, parson AG described the technical content engineer as someone who organizes content “along the content supply chain”, using content management systems, XML, APIs and AI. The tools are newer now, but the job of structuring content for reuse is much the same.

GOV.UK is a public example of this work. Its publishing platform has more than 80 content schemas, from answers to worldwide organizations. Each one is “a JSON Schema that defines the underlying structure of a piece of content”, and they’re published openly, so you can read how each type of page is structured.

Here, the work sometimes gets a title of its own, and sometimes it’s part of a UX writing role:


Own the content platform

This is the most engineering-heavy hat: owning the platform that content runs on. That covers content APIs, search and the retrieval systems that let AI tools answer from accurate company knowledge.

People in this role write production code, often with help from AI coding assistants, and content designers are among the users of what they build. Stripe’s docs are a well-known example: Stripe built Markdoc, an open-source tool, to power its public docs.

These 2 postings show the engineering end of this hat:


Where to start

If you want to find out if these responsibilities already exist where you work, or if there’s room to carve out a space, these questions may help:

  • Are any of your voice and tone guidelines checked automatically, by a linter, a plugin or an AI skill?
  • Is anyone on your team writing prompts or evaluation rubrics for AI features?
  • Does your design system define content fields for its components, such as titles, descriptions, error states and empty states? Who maintains them?
  • Who owns the platform behind your help center or docs, and do they take requests from content designers?

If you’d like to try the tooling side yourself, the ux-writing-skill repo linked above is one place to start. It’s open source, so you can read how its standards are written as instructions for an AI tool.

And if your team already works in GitHub, I’ve also written about UX writers merging their own pull requests.

The UX writer who merged a pull request
How we were (quite literally) handed the keys

Closing thoughts

As usual, I have a lot of questions. How will content people and engineers split the work and own what they do, with AI tools changing so fast? Content platform engineering sounds like the frontier right now, but how long will it feel that way before someone (or something) else catches up and takes that edge away too?

When I tested how well Claude drafts product docs, every draft still needed a writer’s review. So what’s a healthy balance? How much are content people expected to take on? Can we learn to build solid systems fast enough to stay relevant?

Let me know what you’re seeing.

Can Claude help you automate your product docs?
I ran five tests. Here's what didn't work and how to do better next time.
READ MORE FROM ME
CTA Image

Ciao 👋 I’m Elisa, an Italian product writer and translator who believes good design is service. This is where I document my work in UX.

Go to the blog

Let’s talk words

Connect with me on LinkedIn or Medium to talk about all things content design, UX writing and localization.

Contact me