MTSTelecom · 2026 · live
Fixes the text and explains every edit.
Task: Company texts were edited by hand through an editor → Solution: built an AI tool that edits texts to the editorial policy (Tone of Voice) and explains every change → Result: made for different internal departments of the company, with sign-in by corporate email
A Design Center employee pastes a push notification, app screen or email and gets it rewritten to MTS editorial policy and Tone of Voice. Each edit cites a company rule.
Led the project and built the product: a rule base drawn from the company's own documents, the checking logic, a quality test bench, the interface and staff sign-in.
One text, before and after
A test text from our bench, translated from Russian: 12 words of shouting- ONLY TODAY! Don't miss your chance!
- Connect right now or you'll regret it!
- Hurry!!!
Only today! Connect and take your chance. Three edits, each explained: the all-caps shouting is gone, so are the extra exclamation marks and the pressure of "or you'll regret it".
Why you can trust each edit
Testers caught the model inventing reasons. In the new version the rule base has a second layer: edits follow condensed rules, each explained by a verbatim quote from company documents.
The rule base mirrors the editorial policy and Tone of Voice section by section. A script, not a person, builds it from the source documents, so it can't drift.
Every new version runs through a bench of 125 texts. It ships only if it does no worse than the previous one. We rolled back even obvious improvements when results dropped.
What's inside
A push, an app screen, an error, a support reply, a long read.
The company's own rules, not generic advice about good style.
What changed and which rule applied.
125 texts test every new version before it ships.
Only employees with an @mts.ru address.
