Vanja — Multilingual & AI Localization Engineer AI Skill
Instant download · 30-day money-back guarantee. Pay once, keep forever — no subscription. Refund policy
Take the product multilingual: measure per language, control token cost, and roll out in the order the users justify.
- Per-language evaluation with native-speaker review loops
- Tokenizer efficiency and the cost penalty by script
- Glossary and termbase enforcement, locale formatting, RTL
- Cross-lingual retrieval, regional compliance, rollout ordering
Product teams expanding an AI product beyond English who have never measured what quality actually is per language.A multilingual AI engineer charges $125+/hr, this is one file, yours forever.
Drop Vanja into Claude and get a localization engineer who measures quality per language, because English performance never transfers the way the roadmap assumes.
Vanja makes an LLM product work outside English: measuring per-language quality rather than assuming it transfers; tokenizer efficiency and the cost penalty non-Latin scripts pay; language-specific evaluation sets and native-speaker review loops; machine translation versus native generation and when each is right; translation memory and glossary enforcement inside prompts; locale-aware formatting for dates, numbers, currency, addresses and names; right-to-left and bidirectional text; script and encoding pitfalls; retrieval over multilingual corpora including cross-lingual embeddings; culturally appropriate tone and register; regional compliance differences in what a model may say; latency and model availability by region; and a rollout order argued from real user distribution rather than executive preference.
What you get
- →Per-language evaluation with native-speaker review loops
- →Tokenizer efficiency and the cost penalty by script
- →Glossary and termbase enforcement, locale formatting, RTL
- →Cross-lingual retrieval, regional compliance, rollout ordering
How to install
Download the .skill package → open Claude → paste SKILL.md into your Project Instructions or system prompt → describe your requirement → Vanja builds the answer. Includes a full worked example so you see exactly what you get.
# Vanja - Multilingual & AI Localization Engineer You are Vanja, a multilingual and localization engineer. English quality never transfers, so you measure every language separately. ## How you work 1. Build a per-language eval set with native-speaker review 2. Measure tokenizer cost per script before promising a price 3. Decide translation versus native generation per surface 4. Enforce glossary and locale formatting; order rollout by real usage Never ship a language you have not evaluated with a native speaker, and never let an LLM-as-judge score a language it is weak in.
Excerpt from the actual file you'll download.
Four steps. Any AI chat.
- 01Download the file
After checkout, the download link lands in your inbox. Save the file anywhere on your device.
- 02Open your AI chat
Claude, ChatGPT, Gemini, Grok, or Copilot — whichever one you already use.
- 03Paste the file contents
Drop it into the system prompt, Project instructions, or custom instructions field.
- 04Start working
Your AI is now configured as a specialist. Ask it anything inside its domain.
No technical knowledge required. No subscription. Pay once, keep forever.
Works with every major AI chat.
Drop the file into your AI's system prompt, Project instructions, or custom instructions. No setup. No code. No vendor lock-in.
- Claude
- ChatGPT
- Gemini
- Grok
- Copilot
Works with any AI chat that accepts a system prompt or custom instructions.
Ready to specialise your AI?
One drop-in file. Pay once, keep forever — works with Claude & ChatGPT.