gheero team•25 Sep 2026•12 min read

We Thought 500 Hours Was a Number. It Wasn't.

When I first joined Leyu, I wanted to grow in my career and lead complex projects. What I didn't expect was how an Amharic language data-collection initiative across five cities would teach me what it truly takes to implement large-scale projects—and why 550.22 hours deserved three celebrations.

gheero team•09 Apr 2026•7 min read

A Curated Resource Guide to Ethiopian Language Datasets for AI

If you ask a modern AI system to speak Amharic today, it will probably impress you. But if you look a little closer, something feels off. This guide explores the fragmented landscape of Ethiopian language datasets and what it takes to build AI that truly understands how people speak.

gheero team•17 Oct 2024•5 min read

Data Collection With Purpose: The Leyu Approach Local Annotators

Leyu is a platform designed to collect high-quality text and voice datasets for low-resource Ethiopian languages, starting with Amharic, Afaan Oromo, Tigrinya, Af-Somali, and Sidama. These datasets will support the development of technologies that improve digital access for underserved communities—though the mission goes beyond just technical progress.

gheero team•17 Oct 2024•5 min read

We Thought 500 Hours Was a Number. It Wasn't.

A Curated Resource Guide to Ethiopian Language Datasets for AI

Data Collection With Purpose: The Leyu Approach Local Annotators

Leyu And CoGAP: Creating Job Opportunities Through Ethical Language Data Collection

Progress Of Natural Language Processing (NLP) For Ethiopian Languages

Leyu And CoGAP: Creating Job Opportunities Through Ethical Language Data Collection

Progress Of Natural Language Processing (NLP) For Ethiopian Languages - Part Two