Launch HN: Quetzal (YC S24) – Stripe for Internationalization
Quetzal, founded by John and Brendan, simplifies software translation using LLMs, automating string identification and ensuring rapid, accurate translations. They currently pilot for Next.js and seek feedback on challenges.
Quetzal, founded by John and Brendan, aims to simplify software translation using large language models (LLMs). They recognize the challenges of internationalization, which often involves tedious manual processes and can delay product launches. Their experiences at Slack and a retail startup highlighted the inefficiencies of traditional translation methods, which can lead to inconsistent and context-less translations. To address these issues, Quetzal employs a custom Babel plugin that automatically identifies user-facing strings needing translation, thus eliminating the need for manual searches. During the build process, the plugin provides context to LLMs, ensuring accurate translations based on usage scenarios. This approach allows for rapid translation, taking only seconds compared to the days required for human translators. Quetzal's solution not only enhances translation quality but also streamlines the integration of new strings into the codebase. They currently offer a pilot program for Next.js projects and are open to expanding support for other frameworks. The founders invite feedback on current translation challenges and are interested in exploring how businesses can effectively reach new markets.
- Quetzal simplifies software translation using LLMs to improve efficiency and accuracy.
- The platform automates the identification of strings needing translation, reducing manual effort.
- Translations are generated quickly, taking seconds instead of days, ensuring timely updates.
- Quetzal is currently piloting its solution for Next.js projects and is open to supporting other frameworks.
- The founders seek feedback on translation challenges and market expansion strategies.
Related
AI-powered conversion from Enzyme to React Testing Library
Slack engineers transitioned from Enzyme to React Testing Library due to React 18 compatibility issues. They used AST transformations and LLMs for automated conversion, achieving an 80% success rate.
Show HN: I coded my own JSON translation tool to easily localize my side project
Quicklang simplifies global expansion for developers by translating JSON files into multiple languages swiftly. It automates, validates, and organizes translations, offering flexible pricing and refund options. Users praise its time-saving efficiency.
The official Elixir Language Server team
The Elixir programming language has formed a Language Server team to enhance code intelligence tools, consolidate existing implementations, and seek community support, with funding from Fly.io and Livebook.
Show HN: Relari – Auto Prompt Optimizer as Lightweight Alternative to Finetuning
Relari has introduced an Auto Prompt Optimizer to improve language model performance for specific tasks, offering transparency and ease of use, with future features planned and user feedback encouraged.
An update on Llama adoption
Llama, Meta's language model, has surpassed 350 million downloads, with significant growth in usage and adoption among major companies, driven by its open-source nature and recent enhancements.
- Users express interest in the automatic translation setup during the build process, with requests for more information on this feature.
- There are inquiries about support for various frameworks, including Svelte and Vue, as well as plans for mobile app integration.
- Some commenters highlight the importance of context in translations, particularly for languages with formal and informal distinctions.
- Concerns are raised about the challenges of change control in translation processes, emphasizing the need for robust systems to manage updates.
- Overall, there is enthusiasm for the product, with users eager to test it and provide feedback.
16 commentsOf course, we all know that this is very rarely how projects end up getting setup especially in the early stages, and then it's just massive amounts of work to go back and set it up later.
The thing that's the most intriguing to me about what you're describing is automatically setting up translations in the build step where you auto-detect strings to translate. But looking at the site, most of it seems to be focused around the VSCode extension which will just sort of find and replace strings in the source code with t() tags.
Can you talk more about the translations in the build step? Is there a reason you're not talking more about that on the site? (Is it just newer, not very reliable/good, or...)?
The idea that I could just throw something like this into my project, not have t() tags in my source code but still get translations, sounds like magic and I think it would be really neat.
Any plans to extend this to iOS/Android development in the future? I assume it would already be easy to integrate this into React Native.
Also, is there a way for me to provide explicit additional context to the `t` function for the translation? Essentially a string that is appended to the LLM input for translation. For example, in Japanese there is often a significant difference between formal and informal language, and it is common to add post-positional particles such as や, が, and の to make titles and labels sound more natural. I see you have addressed many other special cases around numbers/dates/etc, so certain flags like formal/informal, regional dialect, etc may be valuable future additions.
Overall looks really nice and I look forward to trying this the next time the need arises.
Does your result live update the strings in place if the device locale is changed?
Do you have any method for getting feedback from UI tests? I don’t now, but that is absolutely a feature I was used to previously. We used to OCR off expected areas to ensure things fit etc.
No pricing to be found in the header for an AI product (which you'd expect to be on the pricier side) isn't great either.
The idea of parsing source code to auto inject translations, especially while leveraging machine translations comes up every 2 months.
It’s not solving the problem.
The problem to be solved is change control. Doing translations is (surprise!) cheap compared to controlling changes. Changes referring to the marketing copy changed, the button label changed, a new screen has been added, etc. It needs one system that can track and control changes across apps, translations, files.
If change control is solved, localization boils down to managing CI/CD pipelines.
Related
AI-powered conversion from Enzyme to React Testing Library
Slack engineers transitioned from Enzyme to React Testing Library due to React 18 compatibility issues. They used AST transformations and LLMs for automated conversion, achieving an 80% success rate.
Show HN: I coded my own JSON translation tool to easily localize my side project
Quicklang simplifies global expansion for developers by translating JSON files into multiple languages swiftly. It automates, validates, and organizes translations, offering flexible pricing and refund options. Users praise its time-saving efficiency.
The official Elixir Language Server team
The Elixir programming language has formed a Language Server team to enhance code intelligence tools, consolidate existing implementations, and seek community support, with funding from Fly.io and Livebook.
Show HN: Relari – Auto Prompt Optimizer as Lightweight Alternative to Finetuning
Relari has introduced an Auto Prompt Optimizer to improve language model performance for specific tasks, offering transparency and ease of use, with future features planned and user feedback encouraged.
An update on Llama adoption
Llama, Meta's language model, has surpassed 350 million downloads, with significant growth in usage and adoption among major companies, driven by its open-source nature and recent enhancements.