# AI and Automation

Strings reach Crowdin one at a time, sorted by file, without the screen they came from. Translators see them that way, and so do AI models. A screenshot gives that screen back: upload one, and Crowdin tags every string visible on it. The new generate-screenshots skill automates screenshots for localization: your AI coding agent takes them for you – across web, mobile, and desktop apps – and keeps them current after every release.

This October, Crowdin Assistant (formerly Copilot) started working on its own: automations run your instructions on a schedule, on Crowdin events, or from a webhook. AI agents can now also run your integration apps, so one chat message can bring content from a connected platform into Crowdin, have it translated, and send the translations back. Crowdin Enterprise got a redesigned interface that's easier to use, and a new skill lets your coding agent capture a screenshot for every string, so translators see where it appears. The Editor, reports, and apps also got improvements.

To translate user-generated content without losing your terminology, send it through Crowdin's AI translation API with the translation memories (TMs), glossaries and style guides you already maintain. There are two calls – AI Translate Strings for text and AI File Translations for files – and neither needs a Crowdin project. Reviews, comments, listings and support messages are content no one would ever send to a translator. There's too much of it, and it has to appear in minutes. So it usually goes straight to a machine translation engine that has never seen your glossary. This guide shows how to translate it with the same data your localization team already uses for your product.

By Willian Magalhães, Localization Specialist & Conversation Designer Every Monday I had the same small problem. New content had landed over the weekend, and somebody had to turn it into tasks. Creating the tasks takes maybe five minutes. Figuring out who should get each one is what used to eat my morning. In a TMS you sometimes run into fairly complex cases, and out of the box there is often no UI to handle them comfortably. This is one of them. "Which of my translators has room for this?" sounds like a simple question, but the answer is spread across every project you have. I have used a lot of TMSs and they all have good screens. None of them had a screen for how my team works, and honestly I don't expect one to, because no product can ship a screen for every team out there. So I did what most of us do. I exported a spreadsheet, or asked an engineer to write me something with the API, or I just opened project after project and kept the whole picture in my head. I really believe your problem should be solved the way your team works, not the way your TMS happens to have UI for. And this is where Crowdin is at its best for me. There's the whole apps infrastructure to build on, plus Crowdin Assistant (formerly Copilot), which lets you run localization from a chat. And the chat builds a dynamic UI on the fly. That last part I didn't expect, so I recorded a video to test it on a setup that looks like a real company.

This September, AI took on more of the work in Crowdin, and your organization's context now reaches the teams that translate outside it. Crowdin Assistant (formerly Copilot) runs scripts and builds apps without a developer, AI Pipeline can leave proofreaders only the ambiguous strings, one glossary and TM can cover every project, and teams that translate in their repository get the same Crowdin context.

This August was about giving your AI agent the whole Crowdin toolkit. The MCP Server already lets an agent access your projects, now it has the tools to work in your codebase too. Crowdin CLI 5.0 is a complete rewrite that starts in milliseconds and speaks JSON and TOON. Four new Crowdin Skills came with it, including i18n-setup, which takes an app from hardcoded strings to multilingual on its own. Crowdin Assistant (formerly Copilot) can now build an AI Pipeline from a plain-language description, and you can add to a request while it's still going. There are changes across the rest of the product too: a much faster Editor and file manager on large projects, the first round of accessibility improvements, a Variables tab in the Figma plugin. We also shipped an Adobe Experience Manager connector. Let's dive into the details.

This July, we shipped a new advisor that flags translations gone stale after a glossary or style guide change, opened up serverless apps so anyone can build and host their own tools on Crowdin without a backend, and taught Crowdin Assistant (formerly Copilot) to learn skills specific to your organization. The AI usage report now shows what every token was spent on, by which app, and what it cost. There's also a floating screenshot viewer in the Editor, scope tracking for tasks, noticeably faster loading for large projects, and a batch of Figma plugin updates. Let's dig in.

If you have ever managed a localization project with tens of thousands of strings, you know the struggle: finding duplicate keys, locating untranslated strings with specific tags, or fetching clean Translation Memory (TM) segments manually takes hours. While Crowdin provides powerful querying tools like CroQL (Crowdin Query Language) and GraphQL, crafting precise query expressions isn’t always intuitive. What if you could ask an AI coding assistant (like Claude Desktop, Cursor, or GitHub Copilot) to do that for you? In our previous guide on automating i18n context with AI agents, we covered how AI can automatically generate context descriptions for translators. But AI agents can do much more than just enrich JSON files. By combining AI Agents with Crowdin’s open-source croql and graphql agentic skills, you can also turn natural language questions into valid, optimized Crowdin API queries in seconds. Here is how to set it up and leverage it in your automated localization workflows.

This June, we released Crowdin Dreams, that runs quietly in the background to analyze how your translators edit AI suggestions, then automatically improves glossaries, style guides, and AI prompts without any manual effort. We've also added a new glossary creation skill to Crowdin Assistant (formerly Copilot), letting you build termbases instantly from URLs or files, while the new Cursor app brings real code context into the Editor sidebar so translators never have to guess what a string actually means. For teams managing AI costs, we've introduced spending limits at the organization and user levels, plus a Logs tab that tracks every AI call for 30 days. Read this article to learn more about this and other updates.

Short, ambiguous UI strings like "Name", "Post", or "Run" are the biggest problem in localization. Is "Post" a button to publish content or a forum message? Without context, translators and AI models are forced to guess, which leads to broken UI translations and endless QA review cycles. Instead of manually typing out context descriptions for thousands of keys, you can now delegate this entire process to your local AI Agent (like Cursor, Copilot, or Claude Code). By leveraging the new Crowdin CLI context management commands and Crowdin Skills, your AI agent can automatically download your translation keys, analyze your codebase to understand what those keys do, generate precise context descriptions, and push them back to Crowdin. Based on the core workflow introduced in Context Enrichment with AI Agents, this step-by-step guide will show you how to configure and deploy this automated, agent-driven setup directly in your local terminal. <Aside title="Before automating context"> To ensure your short strings, variables, and plurals are properly captured on the AST level and sent to Crowdin via clean PO files, check out our practical guide on how to extract i18n keys and translate them via CLI using LinguiJS. </Aside>

Bejo Zaden is a leading international name in the breeding, production, and sale of high-quality vegetable seeds. Founded by Dutch pioneer families in 1978, the company has grown into an international family business. Today, Bejo employs over 2,300 people worldwide, operates through 31 international subsidiaries, and proudly sells its seeds in more than 100 countries. Managing this extensive network requires coordinating numerous localization projects within Crowdin and synchronizing content across more than 25 regional languages. To learn more about this workflow, we spoke with Jorrit Stroet, Product Owner at Bejo Zaden, who walked us through their transition from legacy processes to a modern, cloud-based localization setup.

Every two weeks, Holafly sends a new version of its mobile app out to users worldwide. The app helps travelers track data usage, manage roaming, and earn rewards like "HolaCoins" across 16 languages. Behind this global operation is Marcel Pintó Biescas, Head of Mobile at Holafly. We came across his post on LinkedIn, where he shared Holafly’s new localization workflow, and decided to ask more deeply about their approach. Marcel’s team ran into a classic problem: their localization process was slowing them down. This is the technical story of how they fixed it by building a custom AI development skill and connecting it to Crowdin.

When you are running dozens of localization projects, keeping track of every moving piece is hard. How do you ensure your AI engines always have enough context, your glossaries stay updated, and your published content doesn't go stale? This month, we are introducing automated guardrails to handle that operational maintenance for you. Crowdin Advisors: A background check for your project setup that automatically flags missing screenshots, thin glossaries, or unattached style guides, and many more. Re-Translation: A two-click workflow to refresh and upgrade already published AI translations whenever your pipeline improves. Crowdin Assistant (formerly Copilot) in the Global UI: An AI agent pinned directly to your top navigation bar to run complex bulk actions across your workspace via simple chat. Let's dive into the details.

For years, the promise of AI in localization has been shadowed by a frustrating reality: unpredictability. A single monolithic instruction ignores carefully curated glossaries, hallucinates technical tags, or quietly loses a brand's established voice – and you only find out downstream, string by string. That unpredictability is what makes AI translation expensive. Not the tokens – the review. When you can't tell which strings the model got right, someone has to check all of them, and the savings disappear into your reviewers' calendars. AI Pipelines are our answer to that – modular, multi-stage workflows that give an LLM one job at a time and check its work before you ever see it. What you get back is output your team can trust without reading every line of it. Since launch, they have grown into three ways of working: a two-question setup that covers most teams, ready-made templates for common content types, and a fully custom builder for everything else. This guide walks through all three. <ActionBanner title="Sinch publishes 56% of AI translations without human review. See the pipelines behind it." buttonUrl="https://crowdin.com/blog/sinch-localization-with-crowdin-and-undertow" buttonText="Read the case study" buttonId="sinch-case-study-top" />

The headline updates are custom instructions and additional context for AI pre-translation, AI Pipeline presets to help you deploy complex workflows in seconds and major enhancements to our Crowdin plugin for Figma. Beyond this, we’re excited to announce that Crowdin Enterprise is now available on the AWS Marketplace. We’ve also added category-based proofreading rates for better budget control and a highly requested auto-save feature in the Editor. Here’s a closer look at everything we shipped in April.

Snov.io is a sales automation platform used by over 3 million users to find, verify, and engage with leads. Snov.io currently supports 9 languages on its website and 5 languages within its platform, with some high-traffic pages translated into as many as 14 languages. Behind this massive operation is Olena Holovach, Snov.io’s Localization Manager. Starting 6 years ago as a localization pioneer, Olena has transformed the company’s approach to localization into a highly structured, AI-driven workflow.

Every night, while you sleep, your brain does something remarkable. It replays the day to extract signal from noise and form long-term memory. Raw sensory data gets compressed into patterns. Patterns become rules. Rules become the intuition you wake up with. Neuroscientists call this memory consolidation. We call it dreams. Crowdin has been observing recent developments in AI coding tools and we figured out that the very same principle might work for Crowdin. Every time a human editor corrects an AI translation, something valuable is created: information about what kind of translations are acceptable for a particular project. Source text. Machine output. Human decision. Our industry almost always throws this away. We ran experiments, and it seems this is a great way to automate the improvement of expected translation quality.

Bitrefill is the world's largest crypto-only store. Launched in 2014, Bitrefill provides digital gift cards & mobile airtime refills using the Lightning Network technology. Their mission is to turn crypto into everyday purchases. Bitrefill’s products serve people all around the world, and they want everyone to be able to use their services in their native language. Sam Kuritzén, Bitrefill’s Localization Manager, was responsible for modernizing how the company handles translations.

Most AI in localization stops at the string in front of you. It rephrases, it adjusts tone, it flags a QA issue – and then you go back to the dashboards, the filters and the twelve browser tabs to actually run your projects. Crowdin Assistant (formerly Copilot) works the other way round. By bridging the gap between LLMs and the Crowdin API, it executes complex workflows that previously required manual labor, custom scripts, or engineering support. It doesn't just tell you information about your projects, it can act on them.

Turo is the car rental marketplace that is reinventing rental to unlock independence for all. But unlike booking a room, car sharing requires a unique layer of confidence. Guests aren't just looking for a place to stay; they are trusting a vehicle to move them safely through the world. This focus on reliability has fueled Turo's massive growth into a $1 billion business. Today, the company successfully operates across the US, Canada, the UK, France, and Australia. For Turo, localization is the "bare minimum" required to build this essential trust with both hosts and guests. Leading this global expansion is Gideon Hod, Turo's Director of Product Operations. Gideon views his role as the center of a massive feedback loop, bridging the gap between high-level product strategy and the boots-on-the-ground reality of local users. While traditional product management often focuses on top-down roadmaps, Gideon’s approach to Product Ops is built on velocity and execution. Under his direction, Turo sped up their workflow shifting from a manual localization process to a high-speed, AI workflow within Crowdin. <table> <thead> <tr> <th class="align-baseline">SUMMARY</th> <td></td> </tr> <tr> <th>Metric</th> <th>Traditional Workflow</th> <th>AI Workflow</th> <th>Improvement</th> </tr> </thead> <tbody> <tr> <td> <strong>Time to Market</strong> </td> <td>3–4 Months</td> <td>1 Week</td> <td> <strong>90% faster</strong> </td> </tr> <tr> <td> <strong>Total Cost</strong> </td> <td>$10,000+ (Est.)</td> <td>AI Tokens + QA</td> <td> <strong>98% cheaper</strong> </td> </tr> <tr> <td> <strong>Human Intervention</strong> </td> <td>100% Manual Review</td> <td>Near-zero (Fallback only)</td> <td> <strong>Almost 100% AI Translations</strong> </td> </tr> </tbody> </table>

There’s a running joke in the translation industry: two linguists can’t agree on whether a translation is correct. At Crowdin, we don’t think the problem is the linguists – it’s that different humans (and AIs) have different quality standards by default. If requirements are communicated clearly enough, two people and an AI should arrive at the same conclusion. We also see this reflected in the data: the same LLM model, on very similar content, produces wildly different post-edit distances across different clients. This isn't because the model is inconsistent, but because different clients have different expectations. Some require good-enough compliance, while others require strict adherence to specific national spelling rules. Without a shared standard, every variant becomes a potential critical error. While our AI Pipeline and Crowdin Assistant (formerly Copilot) can produce high-quality, coherent translations within a single file, the challenge has always been: How do we guarantee consistency across thousands of files? The answer is Style guides.

Based on original research combining a B2B survey of localization, engineering, product, and security professionals, Reddit community discussions, and publicly available industry data. The question is no longer "whether" but "how". AI translation is no longer an experiment. It has become an operational baseline for enterprise teams managing multilingual content at scale. Based on our original survey of 152 B2B professionals across the US and Canada, roughly 95% of respondents already use AI or machine translation in some capacity – with nearly half doing so frequently and about 18% using it for every translation task. Nearly 9 in 10 require or prefer bring-your-own API keys. Over 91% already have governance frameworks in place or underway. And 1 in 5 report quality incidents since introducing AI translation. But adoption alone tells only part of the story. What matters now is how companies implement AI translation when the stakes include data security, regulatory compliance, brand consistency, and production reliability. The central finding of this research is clear: enterprise teams do not treat AI translation as a plug-and-play capability. They treat it as a managed process – one that demands governance, quality controls, cross-functional oversight, and platform-level orchestration. In other words, the industry has moved past the debate over whether AI translation works. The real conversation is about how to make it work safely, predictably, and at scale.

To work well, AI shouldn't guess; it should gather context. This month, we’re presenting a combination of our already released (AI Pipeline) and new tools (Crowdin Assistant) to eliminate repetitive manual tasks for managers, and provide production-ready translations. The highlight is our new Crowdin Assistant (formerly Copilot). One of its key use cases is analyzing hundreds of translation issues and synthesizing them into a few high-level questions for you to answer. We’re also debuting Dual Preview mode in the Editor to shift from segment-by-segment editing to full document review, and enhancing security in task-based mode.

Our release announcements are getting a makeover. Starting now, we’ll be highlighting some features in "What’s New" summaries, while major updates will now have their own separate posts. This shift ensures we can give you more depth and share important updates the moment they drop. Expect the first of these separate announcements to arrive in just a few days! This February, we released Crowdin CLI 4.14.0, which uses AI and our new Crowdin Skills to pull context directly from your code. This helps translators or AI understand exactly what they’re working on without the guesswork. We’ve also launched the Crowdin Certificates app—a new way to prove your expertise based on your actual work on the platform, not just theory. Beyond these updates, we’ve introduced a Screen Capture extension to make managing screenshots easier and a new traditional menu in the Editor for a more familiar layout. You’ll also find new apps like Langbly MT and Google Private Drive in the Crowdin Store, along with several updates to our Figma plugin.

When Valentina Wainstein joined the localization team at MyHeritage nearly a decade ago, the site focused on helping people find their ancestors. Today, it’s a global AI and DNA powerhouse. It's the kind of company that doesn't just show you a family tree, but helps you unlock your own biology. The company’s growth has been massive following its success in the AI and genetic health space, MyHeritage is now valued at approximately $1 billion. For Valentina, the goal has always been simple: make sure these life-changing discoveries aren't blocked by a language barrier. "Our products can truly change lives," she says, "and no one should miss that opportunity just because they don’t speak English".

OpenAI hype has made many teams believe that a simple API call to GPT-4o is all they need to translate their product. This might work for a simple one-pager or situations where you don’t necessarily need the translation to feel native to your audience. However, there is a big difference between translating a sentence in a chat box and localizing a product at scale. OpenAI is a powerful engine, but it is not a localization platform. It lacks the infrastructure to manage a product. To truly understand why, we need to look at the gap between a raw AI model and a professional localization ecosystem. Besides the comparison, we want to show you how to get the maximum of both worlds: OpenAI and Crowdin.

In the world of fitness, there is a golden rule: If it’s not on Strava, it didn’t happen. With over 150 million users across 185 countries, Strava is the world’s largest digital community for active people. It is the place where professional cyclists and weekend joggers alike go to hunt for Kudos, battle for King of the Mountain titles, and share their latest Strava Art. In 2025, Strava reached a valuation of $2.2 billion in a funding round led by Sequoia Capital. With nearly $500 million in annual recurring revenue, Strava is the top unicorn of the fitness world. We spoke to Eduardo D'Antonio, Globalization Director at Strava, who created a globalization strategy for them. Learn how Strava created a complex localization strategy using Crowdin and went global in just 6 weeks.

This month’s update is headlined by our official n8n integration, which allows you to build custom, no-code automations. We are also introducing "Predicting Ambiguity" to our AI Pipeline to prevent translation errors before they happen, and launching the Corporate Translation Portal to give your entire organization a secure way to get translations that don't require high-quality standards instantly. From expanding our game localization tools for Xbox and Epic Games to rebuilding our Google Drive integration for enterprise-grade security, here is everything we’ve released this January.

A new case study from Crowdin’s eCommerce client – Suitsupply. Suitsupply is an Amsterdam-born menswear brand. As a premium global retailer, Suitsupply operates under a core principle: everything they do must be done the right way – sustainably, ethically, and with their customers at the heart. This commitment to quality extended even to their global content and localization strategy. To understand how Suitsupply overcame its scaling challenges, we spoke to Samer Mikhaeil, E-commerce Operations Specialist at Suitsupply. He explained that the challenge was that their previous translation management system (TMS) couldn't keep pace. It lacked: - automation - integrations - AI capacity to scale their content. This created a bottleneck of slow, manual processes that prevented them from delivering a consistent brand experience across all markets.

We found that there’s not much clear information on AI translation costs. So we decided to make an experiment to find out: How much does it actually cost to translate a real project using AI localization in Crowdin? Important: Often, this price will differ from platform to platform. It's worth noting that our price will be clearly lower. Why? While many platforms add surcharges, Crowdin charges no fees for your AI consumption. You pay exactly the direct API rates charged by providers like OpenAI, Google, and Anthropic. This makes Crowdin the most cost-effective TMS with the widest choice of LLMs. This gives you the freedom to experiment with different AI models and find the best fit for your needs while keeping your expenses at a minimum. <Aside type="tip"> When calculating AI translation costs, remember that different translation management systems have different approaches to pricing. Go and read about the Hosted vs. Processed Pricing Model to choose the best fit for your budget. </Aside>

We’re a week late with this update due to the holiday season, but we’re finally here and excited to share all the great things we’ve been working on. On-Demand AI translation via API for translating dynamic content, security apps for controlled user access, and AI Lip Sync to level up your video dubbing. We’ve also added the new Google Translation LLM integration and various features to improve task management, hopefully saving you some of the manual work.

Jiffy is an online marketplace that connects homeowners and local service pros. It's famous for being fast and reliable, with a 96% perfect rating on over 500,000 completed jobs. Before, Jiffy operated in English only. To expand into the Montreal market, they had to translate their entire platform into high-quality Québec French. Their goal was ambitious: - Launch by Q4 2025. - Stay within their existing headcount and budget. This meant they had to find a way to translate over 200,000 words in 6 months while maintaining perfect quality—all with a single human proofreader. And they did that, using Crowdin, AI, and Figma integration. Let’s dive into the details.

If you have ever tried to use AI for professional translation, you know the drill: you give it a massive list of instructions, and it starts out great... then it gets tired. It forgets your glossary, ignores your style guide, or starts hallucinating. In our latest Crowdin Power Up episode, Dorota Pawlak and Stefan Huyghe looked at why this happens and how a new Crowdin app called AI Pipeline actually solves it by changing the way we talk to AI models.

Nowadays, AI localization is an art that requires more than just linguistic knowledge. As AI becomes an integral part of the localization workflow, mastering the art of crafting effective prompts is important for translators and localization professionals. This article was created in collaboration with professional linguist, localization consultant, and AI trainer Dorota Pawlak. Find out advanced prompting strategies and examples for using Agentic AI in Crowdin to enhance various aspects of the localization process. <Aside type="caution"> We’re saying goodbye to Agentic AI and upgrading everything to Crowdin Assistant. Learn more about Crowdin Assistant features. </Aside>

This month, Crowdin launches tools to make localization faster and more reliable: AI Pipeline, a next-gen AI workflow that self-corrects translations, ensures consistency, and follows glossaries and style guides more reliably; improved Global Search in Crowdin Enterprise; Website Context Viewer for better context when translating headless CMS; Volunteer Certificates to recognize contributions; and platform updates. Plus, new episodes of The Agile Localization Podcast feature Jan Amann (i18n expert, creator of next-intl) on how internationalization shapes product architecture, and Shashi Bhushan (Localization Workflow Strategist) on designing AI-enhanced localization workflows that remain human-centered.

For decades, making your business multilingual meant a slow and expensive process of translating word-for-word. Now, with AI localization, every business can afford to go global. But simply changing the language is not enough; you need to capture the culture, the tone, and the market nuance for every single audience. AI localization is not just a faster version of old-school machine translation. Based on Large Language Models, it produces translations that are tailored to specific local cultures rather than language-to-language interpretation. Read on to discover how AI, combined with human expertise, is making it possible to launch international products faster. Learn what the Crowdin platform is, and which localization tools can help you create AI-powered workflows to optimize your budget. Editor's Note: AI Technology moves fast. Tools that work today may be useless tomorrow. To make sure the content is always up-to-date, we review and update this article every few months with the most relevant insights and newest tools.

This month, we are introducing three significant new features. All three are currently available by invitation only. - Crowdin Apps Vibe Coding: Build Crowdin apps simply by describing them to the AI. - Crowdin Automator: Like Zapier, right in Crowdin, AI-powered. - Video dubbing: a fully fledged AI dubbing tool. Each feature is a new piece of technology that we are actively refining. We would love to hear your thoughts and feedback to help us perfect them and ensure they are useful. We've also added a Website Context Extractor that uses AI to provide your headless CMS text segments with real context from your website, several Crowdin Store app updates, support for Figma Buzz, and more.

In localization, context is king. But what if your most valuable brand context – the nuances, the terminology, the single creative phrases – was locked away? For too long, your AI models couldn't fully use that data. Previously, we've relied on traditional Translation Memory (TM) fuzzy matches, often overlooking anything below 50% or even 10% similarity. Those “tiny matches” contain critical stylistic and terminological clues, but they rarely get displayed or used in our workflows. Crowdin's new Vector Cloud app is here to change that, transforming AI translations, by enabling ai to use the most of available context. In the fourth episode of Crowdin Power Up, host Stefan Huyghe and AI localization expert Dorota Pawlak dived into Vector Cloud. They demonstrated exactly what this is, how to set it up, and how to work with it to transform your AI translation workflows. <ResourcesCTA title="Build an AI Localization Workflow with Crowdin" description={null} class="from-gray-900 to-gray-700 dark:from-gray-800 dark:to-[#0B1918]" titleClass="text-4xl! font-semibold" badge="Free Guide" ctaText="Download Now" ctaHref="https://crowdin.com/learn/ai-localization-workflow-ebook?utmterm=ctacrowdinblog" ctaId="ai-localization-workflow-cta" imgSrc="ai-localization-workflow-books.png" showPodcastLinks={false} />

This September release is packed with updates that make Crowdin more powerful, secure, and adaptable to how teams work today. The highlight is Vector Cloud, a new app that takes AI translations to the next level by grounding them in your own data. We’re also shipping a major upgrade to the Crowdin Plugin for Figma, several security-focused improvements like session timeout settings and stricter API token management, and new ways to support hourly-based payments with time logging and reporting.

Localization often takes a lot of effort, so saving time on manual tasks lets you focus on quality. In a recent Crowdin Power Up episode, host Stefan Huyghe and AI localization expert Dorota Pawlak introduced new features that make your Crowdin workflow smarter and more efficient. Keep reading to discover how these platform updates will: - Eliminate ambiguity by automatically flagging unclear strings. - Handle bulk actions like a pro with simple text commands. - Automate workflows for tasks you perform every day. You'll see how these new tools help translators who are ready to embrace the future of AI-powered localization. Check the full episode on YouTube. <Aside type="caution"> We’re saying goodbye to Agentic AI and upgrading everything to Crowdin Assistant. Learn more about Crowdin Assistant features. </Aside>

In August, we delivered 365 updates and completed 523 tasks. We've presented audio generation and localization with the Crowdin Dubbing Studio. We also continue to enhance our AI tools. Our Agentic AI is now smarter, with new capabilities for editing source texts, adding and managing comments, while the Context Harvester now gathers context with much more accuracy. This release features new integrations with leading AI models and editor updates, including a new UI for localizing non-text file formats (e.g., images and videos), a new way to switch between files, and more.

Model Context Protocol (MCP) is a technology that allows you to connect large language models (LLMs) to external, dynamic data sources and tools to provide more contextual information. Unlike Retrieve Augmented Generation (RAG), which accesses static documents, MCP connects to live resources that change constantly, such as code repositories or content management systems. This dynamic access helps LLMs provide more accurate and context-rich outputs. MCP usage can reduce "hallucinations" in which the AI generates incorrect information.

What if there was a way to make AI in localization not just faster and more cost-effective, but also smarter? Enter the Agentic AI era, which transforms how we approach translation and localization. <Aside type="caution"> We’re saying goodbye to Agentic AI and upgrading everything to Crowdin Assistant. Learn more about Crowdin Assistant features. </Aside> In this article, we'll explore: - What Agentic AI is. - How it differs from earlier AI versions. - How Agentic AI works. - Its impact on the localization industry. - Crowdin's approach to Agentic AI (features, examples, and how it helps translators).

In July, we delivered 441 updates and completed 249 tasks. Agentic AI received custom modes, multi-tab support for handling multiple requests simultaneously, a Max Mode toggle for deeper information processing, and a fullscreen tools. Native AI configuration was simplified with clearer prompts and better shortcut management. Pre-translate now lets you choose any project language as the source, exceptionally useful when translating to multiple closely related languages and dialects via AI. A new QA check flags outdated translations where source text changed but translations didn’t. The editor now supports selecting multiple files for batch actions. We’ve now got over 700 apps in the Crowdin Store, and a few more updates to go with it.

In June, we closed 317 tasks and released 384 updates across Crowdin Enterprise and Crowdin.com. We added new AI models, improved prompts for Source Text Review, introduced Max Mode for Agentic AI, made the editor sidebar customizable, updated Crowdin Store apps, extended CLI functionality, and launched instant fallback in Website Translator using DeepL.

In May, we closed 192 tasks and released 430 updates across Crowdin Enterprise and Crowdin.com. The highlight of the month was Agentic AI – a new autonomous system for translation assistance, inspired by vibe coding. It plans, searches, and validates translations like a real team member. In the Crowdin Store, 7 apps were updated, including major improvements to Webflow, Shopify, DatoCMS, and Bilingual DOCX Export. Pre-translation just got faster, filters got more flexible, and AI got more context-aware than ever. Let’s take a closer look.

In April, we closed 209 tasks and shipped 381 updates across Crowdin Enterprise and Crowdin.com. We introduced a new layout for the multilingual grid view in the Editor and made AI pre-translation faster and more context-aware. Task creation now supports automatic cost estimate reports, and we’ve added filtering and ignore options for integrations, useful if you have lots of content in your CMS. The Crowdin Store also saw multiple improvements, including new export formats and enhancements to existing tools.

This month, we’ve successfully closed 283 tasks and deployed 375 updates. In this article, we’ll walk you through the latest features and enhancements, including the new AI-powered Context Evaluation Report, improved user management with group filters, and exciting updates to pre-translation and reporting. We’ve also added several new integrations, including ABBYY Vantage OCR, Prismic, Synthesia, Braze Campaigns & Canvas Translation, and updates to Webflow, Intercom, Slack, and Jira.

Localization is an ever-evolving field, and staying ahead requires innovation, adaptability, and an acute understanding of both technology and linguistics. In a recent episode of The Agile Localization Podcast, host Stefan Huyghe sat down with Felix Bartz and Olli Gurtschmann, Co-founders of Wordcrafts, to discuss how their company is redefining localization, particularly for the Apple ecosystem.

In February, we closed 238 tasks and completed 446 deployments and introduced AI-powered QA checks, added new AI provider integrations, and improved project management tools. Now, you can automate translation quality checks, use Grok, DeepSeek, and WatsonX as AI providers, and work with more flexible import/export options. We also released updates for integrations like Zendesk, Webflow, and Storyblok, along with improvements to project duplication and branch merging. Here’s a breakdown of everything new.

Localization has never been more essential or complex than it is today. Businesses aiming for global reach rely on seamless translations, localized interfaces, and culturally appropriate content to connect with diverse markets. However, despite all the advancements in localization tools, a critical gap remains: they often fail to meet the needs of buyers. Localization at scale isn't for the faint of heart. With 24 languages, 2.7 million words annually, and over 400 developers producing content daily, Pipedrive has mastered the art of agile localization. At the helm of this monumental task is David Edwards, Group Design Manager of Internationalisation. In a recent episode of The Agile Localization Podcast, David shared insights into how Pipedrive manages its vast and complex localization operations. From the importance of automation to working seamlessly with developers, here's a behind-the-scenes look at Pipedrive's localization machine. Listen to the new episode on: - Apple Podcasts - Spotify - YouTube

In January 2025, we closed 262 tasks and completed 462 deployments. Among the key updates, we introduced Crowdin.ai in beta, offering AI-powered localization aimed at indie developers, focusing on faster turnarounds. Additionally, the new Vector Memory App helps improve translation accuracy by using non-bilingual, relevant, translated files as reference. We also enhanced our Translation Costs and Cost Estimate reports by adding weighted word counts and more.

How do you manage multilingual content across ten languages, process over 6 million words annually, and still keep stress levels low? That's a question Edwin Trebels of the Philadelphia Church of God seems to have mastered. In a recent episode of The Agile Localization Podcast, Edwin joined host Stefan Huyghe to share how his team leverages automation, collaboration, and cutting-edge tools like knowledge graphs to streamline workflows and maintain a culture of transparency. Listen to this episode of The Agile Localization Podcast on: - Apple Podcasts - Spotify - YouTube

Localization has never been more essential or complex than it is today. Businesses aiming for global reach rely on seamless translations, localized interfaces, and culturally appropriate content to connect with diverse markets. However, despite all the advancements in localization tools, a critical gap remains: they often fail to meet the needs of buyers. In this episode of The Agile Localization Podcast, host Stefan Huyghe is joined by Localization Consultant Boryana Nenova. They discuss the importance of shifting localization tools from translator-centric to buyer-centric, or at the very least, to be appropriate for both. Boryana dives into top features buyers demand in localization tools, why developers and linguists are clashing over localization tools, AI’s localization impact, and what localization tools are missing. Listen to this episode of The Agile Localization Podcast on: - Apple Podcasts - Spotify - YouTube

In the last release notes of the year, we're going to share exciting updates we did this month. In December 2024, our team closed 214 tasks from the task manager and completed 383 deployments. It includes the ability to add AI proofreading as a workflow step in Crowdin Enterprise, reusable custom properties fields for AI prompts, and the ability to share limited access to integrations with ZenMode. We’ve also enhanced screenshot auto-tagging, brought pre-translation progress tracking on the project page, and introduced a free course to help linguists get started with Crowdin. Dive in to explore all the updates. <YouTubeEmbed id="xY3g37JKHJo" />

Polhus, used Crowdin’s AI and DatoCMS integration to localize their website. 75% of translations were AI-generated and ready for publication, saving $80,000 and hours of work. With AI and human review, they achieved high-quality translations for 1.6 million words in 7 languages. Curious about how they did it? Read the full story in this post.

In November 2024 our team closed 246 tasks from the task manager and completed 398 deployments. This month, we’ve rolled out several new features and improvements aimed at making your work at Crowdin better. You can now tag and manage strings meta information directly from the Screenshots tab, get AI-suggested metadata in the Simple Term Extractor app, and use Backtranslation app to check translation accuracy even when you don't speak the target language. We’ve also introduced an automated screenshot capturing for Web and Mobile, better AI proofreading, enhanced LQA annotations in the LQA app, and more. Let’s explore these updates in detail. <YouTubeEmbed id="5Lw85urmQKc" />

This month, we’re rolling out some powerful updates. From a revamped Context Harvester CLI with a context quality check to new Crowdin Store apps like Language Fallback Configurator and an AI debugging tool. For complex projects, we’ve introduced cross-review approval for tasks and improved savings visualization. Plus, discover multilingual terminology extraction with AI that continuously refines your glossary across languages as your project is translated. One AI model builds a multilingual glossary, while another uses this glossary to pre-translate. In October alone, our customers used over 10 billion AI tokens (11.6 billion, to be exact) when our team closed 517 tasks from the task manager and deployed 475 times. Enjoy checking out the latest features. <YouTubeEmbed id="53Suiz9y5ok" />

Lightning fast, continuous localization. Decent quality. Super affordable. All automated with no humans in the loop. This guide shows you how to build a multilingual product on autopilot using Crowdin. The workflow is tech stack agnostic and suitable for web, mobile and desktop applications, as well as their documentation and marketing materials. Let AI to have all the information it needs to produce the best possible AI translations. You still are able to bring in human linguists to improve translations. Prerequisites: You have an internationalized web or mobile application. This means that all translatable texts from the UI have been extracted into the resource files. The implementation time of this guide is 30 minutes. <BlogCTA title="Looking for a Complete General Guide?" buttonUrl="https://crowdin.com/blog/ai-localization" buttonText="Read More About AI Localization" buttonId="read-more" />

In September 2024, we deployed 360 times and closed 265 tickets from our task manager, including bug fixes and improvements. Below is the selection of the new things we’re most excited about, like fine-tuning results visualization, the ability to fine-tune models based on TM data, the AI Automator app, updated Crowdin Docs website and developers portal, and more. <YouTubeEmbed id="130WW2nYFI" />

In August 2024, we deployed 170 times and closed 340 tickets from our task manager, including bug fixes and improvements. Below is the selection of the new things we’re most excited about like our first AI Agent for Proofreading, AI QA checks, and more. <YouTubeEmbed id="TsmRpqrlx24" />

Welcome to the July 2024 Crowdin update. This month, we've added the Pre-Translation Efficiency Report, improved Amazon and DeepL MT integrations, and introduced the AI Token Usage Report. You'll also find support for new AI models, new QA checks for subtitles, and the ability to set default TM and glossaries for projects. Plus, explore new apps in the Crowdin Store. Read on for the details.

In June, we introduced screenshot-enhanced translation using multimodal AI models such as GPT-4, potentially improving translation quality for UI projects. We also introduced automatic AI pre-translation and revamped AI management features, including sample prompts and a simplified UI. Improved user management in Crowdin Enterprise now supports bulk operations and detailed role tracking. New APIs and EU data residency capabilities increase flexibility and compliance.

Dive into advanced AI translation insights, showcasing up to 96% accuracy with optimized prompts. Welcome Mistral AI, our new AI assistant provider, and explore seamless translation and proofreading tasks. Experience Crowdin Enterprise's new dark theme for a more comfortable user experience, and boost your localization workflow with custom spellchecks, and fields.

From enhanced pre-translation options and AI content checks to updated plugins and API enhancements, this update is packed with features to improve your localization workflow. Let's dive into what's new at Crowdin.

Welcome to Crowdin's December 2023 update. In this edition, we're excited to announce the Google Gemini AI provider, along with other major new features and the introduction of the Crowdin AI Assistant's new features. The new Strings projects are now accessible on crowdin.com. Check out the latest features, improvements, and plugin updates below.

In this update, Crowdin introduces features enhancing translation accuracy, project management transparency and important AI Assistant updates. Crowdin Enterprise now offers an auto-approve option for streamlined content approval, and Crowdin preserves approval status for strings in Translation Memory. The Screenshots Preview app simplifies UI copy translation when additions like Wix CSV and API updates contribute to a more efficient localization workflow.

If you're curious about how Crowdin AI can enhance your localization projects, this article answers some frequently asked questions about price, customization, and best practices. Let’s discover if an AI can translate and the best way to use AI tools for language translation. <YouTubeEmbed id="ZqfvNIMpp4Q" />

It's fascinating to watch how LLMs are taking over the translation industry. In the last few months, we've seen many great applications of how machines can assist and perform translation tasks. The reason LLM's seem to be taking over is because of the context they can accept and leverage when translating. Today, we're very excited to announce immediate access to the latest version of Crowdin AI with the fine-tuning feature! A way to feed TM's and glossaries of any size before asking the LLM to translate your content. Our complete solution for AI localization.

This month, much like in the recent past, our focus has been dedicated to implementing AI into localization workflows. As we continue to explore the potential and possibilities that artificial intelligence brings, it's worth mentioning other exciting updates as well.

The first experiments with machine translation began in the 1950s. At that time, the technology was more like an automated dictionary. Machine translation became more useful in the 90s when statistical machine translation was introduced. However, the translation quality was still far from human translation. In 2010, when Google introduced Neural Machine Translation, things changed. Machine translation became a useful tool, which in many cases, influenced both the work of professional translators and clients' budgets. Now, AI is actively searching for its place within all industries, and localization is no exception. Let's see how you can run AI pre-translation and the potential of AI in localization.

Crowdin has announced some exciting updates in April 2023. Here's a quick rundown of the new hotkeys, AI Assistant, Facebook Ads, Google Ads apps, and other improvements that have been added to the platform.

Greetings and welcome to the latest edition of What's New at Crowdin for March 2023. We have some exciting updates and new features to share with you that will further enhance your localization experience. This month, we announced Crowdin’s AI chatbot, added 7 new apps to our marketplace, including Google Sheets, BeLazy, Placeholder Manager and more. We've also made some enhancements to the Crowdin Developer Portal, as well as added support for a new NestJS format. Read on to discover more about the latest developments at Crowdin.

Today we are happy to announce the invitation to the beta version of our new app – an AI chatbot built on OpenAI's ChatGPT API. Large language models are the hottest topic around our newsfeeds these days. While companies are still figuring out how to use them most efficiently, it is clear that this technology will change the way people and companies work in the near future. Crowdin has been experimenting with AI and ML for years to offer the most advanced technologies and make sure we enable our users to be most productive while doing global business. If you are looking for a comprehensive overview of these strategies, check out our AI localization guide. AI chatbot built on OpenAI's ChatGPT API we're showing today is just one of the first successful experiments we're excited to share. This is an private beta, and you're welcome to join.