Crowdin’s Post

A few months ago, I went further into implementing 𝗻𝗲𝘄 𝗮𝗻𝗱 𝗺𝗼𝗿𝗲 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝘄𝗮𝘆𝘀 𝘁𝗼 𝘁𝗿𝗮𝗻𝘀𝗹𝗮𝘁𝗲 𝗮𝘀𝘀𝗲𝘁𝘀 𝗮𝗻𝗱 𝘀𝘁𝗿𝗲𝗮𝗺𝗹𝗶𝗻𝗲 𝘄𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀 for Sinch, where I work as #LocalizationManager. We already rely on 𝗲𝘅𝗰𝗲𝗹𝗹𝗲𝗻𝘁 𝘁𝗼𝗼𝗹𝘀 like DeepL, Blackbird.io, and the solutions provided by Smartling, but we wanted to go beyond that. 💡 Together with the amazing Nicola Calabrese from Undertow, we designed the 𝗻𝗲𝘅𝘁 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻 𝗼𝗳 𝘁𝗵𝗲 𝗦𝗶𝗻𝗰𝗵 𝗹𝗼𝗰𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝗽𝗿𝗼𝗴𝗿𝗮𝗺, one that combines Large Language Models (#LLMs) with proven localization technologies. 💻 Our goal: to create a scalable, intelligent system that adapts to content type, language, and business impact. To make that happen, we have: ✅ 𝗗𝗲𝘀𝗶𝗴𝗻𝗲𝗱 𝗮𝘂𝘁𝗼𝗺𝗮𝘁𝗲𝗱 𝘄𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀 (that for example include term extraction and glossary creation with context and equivalent term proposals in our target languages) ✅ 𝗧𝗮𝗶𝗹𝗼𝗿𝗲𝗱 𝗽𝗿𝗼𝗰𝗲𝘀𝘀𝗲𝘀 based on 𝗰𝗼𝗻𝘁𝗲𝗻𝘁 𝘁𝘆𝗽𝗲, 𝘁𝗮𝗿𝗴𝗲𝘁 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲, 𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲 𝗺𝗲𝘁𝗿𝗶𝗰𝘀 (traffic, asset life, etc.)... ✅ 𝗦𝘂𝗰𝗰𝗲𝘀𝘀𝗳𝘂𝗹𝗹𝘆 𝘁𝗲𝘀𝘁𝗲𝗱 𝘁𝗵𝗲 𝘂𝘀𝗲 𝗼𝗳 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗟𝗟𝗠𝘀 𝗮𝘁 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝘀𝘁𝗲𝗽𝘀 ( translator, reviewer, QA assistant, and more) all within the Crowdin AI environment, achieving excellent results (suprisingly for me, the best being using Gemini as both #translator and #reviewer) 💪 The outcome? A smarter, faster, and more cost-effective localization pipeline that brings together the best of AI innovation and traditional localization expertise. ⚙️✨ If you’re curious about how we’re reshaping localization to help companies scale intelligently and sustainably, I’d love to connect, or you can schedule a chat here 👉 https://lnkd.in/dKBhrN34 #Localization #AI #Translation #Innovation #LLMs #Automation #CrowdinAI #SmartGrowth #Sinch