Skip to content
Technical Translation and Software Localization

Ship every release in every language

Software strings, help centers, engineering documentation, e learning and marketing content localized by native speaking linguists with domain training. Translation memory and glossaries keep it consistent; TMS and API connections keep it continuous.

What We Localize

Content types

Technical content fails in translation when the linguist does not understand the product. Our teams are grouped by domain, read your documentation before they start and query anything ambiguous instead of guessing.

Software UI strings and apps

Web, desktop and mobile interfaces, onboarding flows, error messages and notifications. We respect character limits, placeholders, plural rules and gender agreement, and flag strings that need context.

Right to left layout testing

Arabic, Hebrew, Persian and Urdu builds are reviewed in the running product for mirroring, alignment, truncation, bidirectional text and icon direction, with screenshot backed reports for your developers.

Help centers and documentation

Knowledge bases, release notes, API references, admin guides and support macros, kept in sync with the UI terminology so users see the same words in the product and in the article that explains it.

Engineering manuals

Operating and maintenance manuals, safety data sheets, installation guides, specifications and drawings. Units, tolerances and warnings are checked in a dedicated pass. Medical device IFUs are covered under medical translation.

E learning

Course content, quizzes, voice over scripts and on screen text exported from authoring tools, returned in the same structure so your team can rebuild the course without manual copy and paste. Subtitles via our transcription team.

Marketing adaptation

Landing pages, campaigns, app store listings and product messaging adapted for the market rather than translated word for word, with a brief on tone, audience and the claims that must survive intact.

File Formats

Send the native file, get the native file back

We work directly in resource and document formats so nothing is retyped and nothing breaks. Keys, placeholders, variables, inline tags and comments are protected during translation and validated before delivery.

If your format is not listed, send a sample. Most structured formats can be handled through a filter or a short conversion step that we set up once and reuse.

JSON XLIFF 1.2 and 2.0 PO and POT RESX XML YAML Android XML iOS strings and stringsdict CSV and XLSX DOCX and PPTX InDesign IDML Markdown HTML SRT and VTT DITA SCORM exports
Language Assets

Translation memory and glossaries

Both assets belong to you. We build them on the first project, maintain them on every release and hand them over in standard formats whenever you ask.

Translation memory

Every approved segment is stored per client, per language and per product line. On the next release, exact and close matches are reused and reviewed rather than retranslated. The result is faster delivery, lower cost on updates and wording that does not drift between versions.

  • Match analysis with every quote so you see what is new and what is reused
  • Alignment of existing translations to seed the memory on day one
  • Exported as TMX on request

Glossaries and style guides

A termbase of product names, UI labels, approved and forbidden terms and definitions, enforced by automated checks during translation. A short style guide per language covers tone, formality, date and number formats and how to handle untranslatable elements.

  • Built with your product and support teams, then locked for the release
  • Shared with your in country reviewers so feedback is consistent
  • Exported as TBX or spreadsheet on request
Continuous Localization

TMS and API connected workflows

Product teams ship weekly or daily. Emailing spreadsheets cannot keep up. We connect to your translation management system or repository so new and changed strings reach us automatically, translations return through the same channel and your build picks them up without manual handling.

If you do not run a TMS, we can host the workflow on ours and give your team reviewer access, or work from a shared repository branch with a simple pull request cycle.

  1. 1. Connect

    Integration with your TMS, repository or CMS through its API or connector. Source strings are pulled on a schedule or on commit.

  2. 2. Translate

    Translation memory and glossary are applied, a domain trained linguist translates the new content and an independent editor reviews it.

  3. 3. Validate

    Automated checks for placeholders, tags, length limits, numbers and terminology, then linguistic QA in context where a build is available.

  4. 4. Return

    Translated files are pushed back through the same connection, ready for your build. Status is visible to your team throughout.

Quality Assurance

Linguistic QA and in context review

A string that reads well in a spreadsheet can still be wrong on screen. QA happens twice: once on the text and once in the product.

Two tier linguistic review

Translator plus independent editor on every batch, following an ISO 17100 aligned workflow, with automated checks for tags, placeholders, numbers, length and glossary compliance.

In context review

Linguists walk through the staging build or screenshots to catch truncation, wrong context, concatenated strings and layout issues, and log each finding with a screenshot and a suggested fix.

Reviewer feedback loop

Your in country reviewers comment in the TMS or a shared sheet. Accepted changes update the memory and glossary so the same correction never has to be made twice.

Frequently Asked Questions

Localization questions

Which file formats can you work with directly?

We handle JSON, XLIFF, PO, RESX, XML, YAML, CSV, DOCX, XLSX, PPTX, InDesign packages, Markdown, HTML and SRT, among others. You receive the same format back with placeholders, tags and keys intact, ready to commit or import.

What is translation memory and how does it reduce cost?

Translation memory stores every approved sentence pair. When a new version repeats or closely matches earlier content, the match is reused and reviewed instead of translated from scratch, which shortens turnaround and lowers cost on updates.

Can you connect to our translation management system or repository?

Yes. We work inside the major TMS platforms and can connect through their APIs or a repository integration so new strings flow to us automatically and translations return without manual file handling.

Do you test right to left languages in the actual interface?

Yes. For Arabic, Hebrew, Persian and Urdu we review the built product or staging environment to check mirroring, text direction, truncation and mixed direction strings, and report issues with screenshots.

How do you keep terminology consistent across releases and vendors?

We maintain a client glossary with approved terms, forbidden terms and definitions, enforce it through automated checks during translation and share it with your internal reviewers so everyone works from the same source.

Planning a release in new markets?

Send a sample of your resource files or documentation and the target languages. We will return a scoped quote, a match analysis and a proposed workflow that fits your release cycle.