Five services, one goal: software that speaks your users' language.
Everything below is work we do ourselves, with named people. There is no reseller layer and no anonymous vendor pool.
A defined piece of work with a fixed scope — a first locale, an audit, a quality review of an existing translation. You get a written scope, a fixed price and a delivery date before anything starts.
- Scoping call and written proposal within two working days.
- Fixed price; changes are re-quoted rather than absorbed silently.
- All artefacts handed over in open formats at the end.
Continuous localization for a project that ships regularly. We hold reserved capacity each sprint so new strings never sit waiting for a translator.
- Reserved translator and reviewer hours per release cycle.
- CI integration so extraction happens without anyone remembering to do it.
- Quarterly quality report with per-locale scores and trends.
Unfunded free software projects are why this collective exists. We keep capacity aside every quarter for work that no one is paying for.
- At cost or free, depending on the quarter — we will tell you honestly which.
- Priority for projects used by communities that are underserved in their own language.
- We help you build a volunteer team rather than making you dependent on us.
Software Localization
We take your product from a single hard-coded language to a shipping, multi-locale build — extracting strings, wiring up the catalogue format your stack already uses, and running the translation cycle release after release.
- String extraction and catalogue setup for gettext, ICU MessageFormat, Fluent, XLIFF, .resx, .strings and JSON/YAML resources.
- Plural, gender and context rules configured per locale — not a flat key/value dump.
- Continuous localization: every merge to main produces an updated POT and pushes it to the translation platform.
- Locale-aware date, number, currency and address formatting reviewed against CLDR.
- Pseudo-localization builds so layout breakage is caught before a single word is translated.
At a glance
| Deliverable | A merged pull request, a populated catalogue and a release checklist your team owns. |
|---|---|
| Typical timeline | 2–6 weeks for a first locale pass |
| Good fit for | gettext / ICU / Fluent, CI-driven, 40+ locales |
Linguistic Quality Review (LQA)
An independent, scored review of an existing translation — the sort of pass that finds the mistranslated destructive action before a user does, and gives you a defensible number rather than a vague feeling.
- MQM-style error typology: accuracy, fluency, terminology, style, locale conventions, markup.
- Severity weighting so a broken placeholder never scores the same as a comma.
- In-context review against real screenshots and builds, not spreadsheets in isolation.
- Terminology and consistency audit against your glossary and translation memory.
- A written report with per-locale scores, reproducible examples and a prioritised fix list.
At a glance
| Deliverable | A scored quality report plus corrected files ready to merge. |
|---|---|
| Typical timeline | 5–10 working days per locale |
| Good fit for | MQM scorecard, In-context, Per-locale report |
Website & Documentation Localization
Marketing pages, help centres, README files and API references translated with the structure intact — including the technical plumbing that decides whether search engines ever find your translated pages.
- Markdown, reStructuredText, AsciiDoc, MDX and static-site content pipelines.
- hreflang, canonical URLs, localized slugs and sitemap wiring done properly.
- Screenshots and diagrams re-shot in the target locale where the UI is visible.
- Style guide and tone-of-voice adaptation per market, not a literal carry-over.
- Incremental re-translation: only changed paragraphs are re-billed and re-reviewed.
At a glance
| Deliverable | Localized content in your repository, with build passing. |
|---|---|
| Typical timeline | From 1 week for a docs set |
| Good fit for | Docs-as-code, SEO-safe, Incremental |
Translation Team Management
Volunteer translation communities do not run themselves. We recruit, onboard, mentor and retain the people who translate your project — and put the review structure in place that keeps quality steady when contributors come and go.
- Team charters, reviewer ladders and clear escalation paths per language.
- Onboarding docs, glossaries and style guides written so a newcomer is productive on day one.
- Recognition and credit programmes that measurably reduce contributor churn.
- Release coordination: string freeze calendars, translation sprints and deadline comms.
- Conflict handling and code-of-conduct support for language teams.
At a glance
| Deliverable | A staffed, documented language team with named reviewers. |
|---|---|
| Typical timeline | Ongoing, reviewed quarterly |
| Good fit for | Volunteer-first, Reviewer ladder, Retention |
i18n Engineering & Consulting
Before anything can be translated well, it has to be internationalized well. We audit the codebase, fix the structural problems — concatenated strings, baked-in layout, byte-based string handling — and leave the team able to keep it that way.
- Codebase audit for concatenation, hard-coded formats, sorting and casing bugs.
- Bidirectional (RTL) layout support: logical CSS properties, mirroring, bidi isolation.
- Unicode correctness: normalization, grapheme-aware truncation, case folding, collation.
- Lint rules and CI checks so regressions are blocked at review time, not found in production.
- Team workshops and a written playbook so the knowledge stays after we leave.
At a glance
| Deliverable | An audit report, merged fixes and CI rules that hold the line. |
|---|---|
| Typical timeline | 3–8 weeks depending on codebase size |
| Good fit for | Code audit, RTL / bidi, CI guardrails |
Before you write to us
Ready to speak to the whole world?
Tell us about your project and we will come back within two working days with a scope, a timeline and the names of the people who would do the work.