Hi y’all! 👋

It has already been a month since we began the Legesher’s “Explorer Program” (wow, time flies), where we piloted a private version of Legesher with a small group of developers and open source contributors.

our community showed up 🤝

From all over the world, developers started contributing; debugging + trying out new features of Legesher in 26 different languages! As part of the onboarding process to Legesher, we asked what languages they wanted to be supported. Thanks to those that spoke up for their language, we now have 51 languages supported in Legesher!

better, together 🫶

Our greatest strength is the community that pours into Legesher. Their desire to create while unhindered by linguistic boundaries is what uncovers simple yet truly novel solutions. They found bugs 🐛 this insomnious mother of 3 hadn't, contributed better translations than what AI generated, and used it to teach AI models to think in an entirely different way. See some examples below! 🦋

@sarah_j_200 — Urdu keywords 🗝️
S* elevated Legesher's Urdu language pack to a whole new level by creating space for keywords in traditional Urdu, Modern Urdu, and Roman Urdu. This will give access to even more Urdu speakers, helping individuals across different contexts, regions + preferences! (If you speak Urdu, we’re still experimenting with this feature, so would love your feedback!)

@confuged_baji — the Arabic comma 💻
R* found a key bug where Legesher's parser didn't account for the Arabic comma (, vs ،). This is the kind of thing only a native speaker would catch, and why open source contribution is so essential to what we are building!

So, the question beckons: what will YOU build?

🧗🏽‍♀️ Join the Explorer Waitlist Today! 🧗🏽‍♀️

💫 That’s all for now! Stay tuned for our next newsletter where we will share updates regarding our participation in Cohere AI’s open science research initiative, Expedition Tiny Aya

With love,

Madi ✨
Founder of Legesher