The English-Only Problem
Most AI workflow automation tools are English-only. Even the biggest players — Zapier, Make, n8n — offer English UIs with no AI template localization. If your team speaks Traditional Chinese, Japanese, or Spanish, you’re stuck navigating English menus, reading English documentation, and writing English prompts. The implicit message is clear: automation is for English speakers first, everyone else second.
This isn’t just an inconvenience. It’s a barrier to adoption. When a marketing manager in Taipei needs to set up an automated customer reply workflow, they shouldn’t have to mentally translate every label, instruction, and template before they can start working. When a customer support lead in Tokyo configures escalation rules, the tool should speak their language — not force them to speak its language.
We built JieGou differently because our founding team is multilingual. We’ve experienced this friction firsthand. And we believe AI tools should serve everyone, not just the English-speaking world.
What We Built: 7 Locales, End to End
JieGou now has complete internationalization across 7 locales: English, Traditional Chinese (zh-TW), Simplified Chinese (zh-CN), Japanese, German, French, and Spanish. This isn’t a partial translation or a community-contributed language pack bolted on after launch. It’s a ground-up architectural decision that touches every layer of the product.
We replaced 425+ UI strings with Paraglide i18n keys. Every button, label, error message, tooltip, navigation item, and confirmation dialog has been localized. The settings page, the workflow builder, the recipe editor, the governance dashboard — all of it renders in the user’s chosen locale.
But UI translation is table stakes. What makes JieGou different is what happens beneath the interface.
Beyond UI Translation: Locale-Aware AI Templates
This is the real differentiator. When a zh-TW user installs the Marketing Starter Pack, every recipe name, description, input field label, and instruction appears in Traditional Chinese. The AI templates themselves adapt — not just the buttons around them.
Consider what this means in practice. A Japanese user browsing the Customer Support department pack sees recipe titles like “顧客フィードバック分析” instead of “Customer Feedback Analysis.” The template instructions that guide the AI are written in Japanese. The placeholder text in input fields is Japanese. The entire experience is native — not translated after the fact, but designed to work in that language from the start.
This matters because AI templates aren’t just labels. They contain prompts, instructions, and context that shape how the AI behaves. A machine-translated prompt often produces worse AI output than a thoughtfully written native-language prompt. We took the time to get this right.
Why APAC Matters
LINE is the dominant messaging platform in Taiwan, Japan, and Thailand with over 200 million users. For enterprise teams in these markets, LINE isn’t an alternative to Slack or Teams — it’s the primary communication channel with customers, partners, and internal teams.
JieGou’s LINE integration goes deep: group chat support, Flex Messages for rich interactive content, typing indicators for natural conversation flow, and webhook-based real-time message handling. Combined with native zh-TW and ja locale support, JieGou is the only AI automation platform purpose-built for APAC enterprise teams.
No other automation platform offers this combination. Zapier has a basic LINE integration with no group chat support. Make has limited LINE actions. Neither offers a localized UI or locale-aware templates. If you’re an enterprise team in Asia-Pacific, your options have been: use an English-only tool with basic LINE support, or build custom integrations from scratch. JieGou changes that equation.
The Technical Approach
We chose Paraglide for compile-time i18n, which means zero runtime overhead. There’s no client-side translation library adding kilobytes to the bundle, no runtime string lookups slowing down page loads. Translations are resolved at build time and shipped as static strings. The result is a fully localized experience with the same performance as a single-language app.
Our blog content lives in 7 locale directories, each with independently authored posts. The marketing site builds 5,300+ pages across all locales — every feature page, blog post, and documentation page exists in all 7 languages.
On the product side, the pack installer resolves template content per account locale. When you install a department pack, the system checks your account’s locale setting and pulls the appropriate localized version of every recipe and workflow template. This isn’t a find-and-replace on English strings — each locale has purpose-written content.
Critically, we don’t rely on machine translation. All content is human-reviewed. Machine translation might get you 80% of the way there, but the remaining 20% — idiomatic expressions, technical terminology, cultural context — is what determines whether the product feels native or foreign.
What’s Next
We’re expanding entity-level i18n to the remaining locales: zh-CN, ja, de, fr, and es. This means more department packs with native-language templates, more workflow templates written for specific regional use cases, and continued investment in APAC messaging channels.
Our roadmap includes deeper integrations with regional platforms, locale-specific compliance templates for governance requirements that vary by country, and expanded documentation in all supported languages.
The goal is simple: wherever your team is, whatever language they speak, JieGou should feel like it was built for them. Because it was.
Try JieGou in Your Language
If you’ve been waiting for an AI automation platform that speaks your language — not just in the UI, but in the templates, the documentation, and the AI itself — we built this for you.
Get started in your language and see the difference a multilingual-first approach makes.