Anonymous case

Can AI write at your quality level, or is it just fast?

Producing text with AI is trivial. Producing the right text reliably is not. We build content systems for cases where quality, language and substance matter. A wrong sentence is expensive.

Everyone can generate text. Few do it well

The difference is in the quality of the text, in the language and in the substance.

Ten texts, someone reads them. Ten thousand, nobody does. At that scale, review has to be built into the system, or quality drifts until a reader notices.

Why no client names2x10

Our clients hire us to build what gives them an edge. Confidentiality is part of the job. Publishing the details could give that edge away. We’ve delivered in regulated environments and on projects worth hundreds of millions.

What we've built

An internal tool behind one of a client's products, built up over more than a year for that one domain.

One agentic workflow writes, one reviews the substance and can send it back rather than onward, one translates. Tens of thousands of pages in several languages, each one checked before it goes live.

Several independent workflows write the same text without seeing each other's work. Automated checks then decide which one survives: length budgets, forbidden phrasings, typography, brand rules.

Independent drafts beat a revised first draft. When you revise, the blind spot stays where it was.

Also built by 2x10

  • Translation Agent

    Translation into several languages, treated as its own quality step rather than a by-product.

  • Content Agent

    specialist literature evaluated automatically, every finding traceable to its source.

  • Compliance Checker

    content checked against internal guidelines and brand rules, not by sampling.

Production questions2x10

What makes them hold up

Volume is the easy part. Three questions decide whether what comes out is publishable.

  • Can the system say no?A reviewer that is allowed to reject, and routing that knows when a topic is too thinly supported to be written confidently.
  • What does one unit cost?Without control the invoice grows faster than the content does, and it shows up a month too late.
  • How is quality measured?What isn’t checked drifts with every iteration, quietly, until someone reads a page and asks who approved it.

What we could build for you

  • We produce content at volume with the expert check built into the machine.
  • We review, update and translate what you already published, rather than leaving it to age.
  • We enforce your brand voice and your internal rules automatically instead of describing them in a document.
  • We turn sources into publishable material, with every statement traceable to where it came from.
  • And whatever you have in mind that AI and solid software engineering could make possible.

Under the hood

Context engineering · source retrieval · evidence stratification · domain knowledge bases · Evals · automated subject review · guideline and brand checks · human-in-the-loop · Orchestration · structured outputs · translation workflows · cost control per unit

Most engagements start small: talking it through, a few weeks of engineering, something real you can judge. If it works, we take it further.If it doesn’t, you find out early

More work2x10

N° 06 Contact

Tell me what you’re planning

No form, no ticket system. Email me directly and you get me, not an inbox someone else manages.

Moritz MiedlerMoritz Miedler · Founder, replies personally
2x10 Technologies · Viennahello@2x10.com