CultureAugust 14, 2026 · 5 min read·MenteE AI Team

Building MenteE AI in 2026: From Idea to Product Lab at menteeai.org

How MenteE AI at menteeai.org became a product lab in 2026 — building mentee-embed-v1, shipping AI platforms, and publishing open research at Zenodo.

CultureMenteE AIBuilding in Public

MenteE AI (menteeai.org) started as a simple bet: stop selling services by the hour and start building systems that run. In 2026 that bet became a product lab that ships.

From prototype to product lab

The early work was engineering-led — full-stack, AI, shipping. The turning point was choosing to build platforms, not decks. The domain menteeai.org reflects that focus: a lab measured by production uptime, not presentation slides.

First proof: mentee-embed-v1

We chose a hard first demo: trilingual embeddings for Arabic, English and Urdu from absolute zero on a single GPU. Published at doi:10.5281/zenodo.22087139 with an honest benchmark report at /research — avg MRR@10 0.585 beating MiniLM-L6-v2 and cross-lingual EN↔UR 0.757 where many small models fail.

Building in public

MenteE is remote-first, product-focused and open by default. Code at github.com/MenteE-s, model at huggingface.co/MenteEAI/mentee-embed-v1, and open roles at /careers. We ship, we publish, we iterate.

MA
MenteE AI Team

Author · MenteE AI — menteeai.org · syab.tech

Cite mentee-embed-v1: Shah et al. (2026). mentee-embed: Training Competitive Multilingual Text Embeddings from Scratch for Arabic, English, and Urdu. Zenodo. doi:10.5281/zenodo.22087139 · Technical Report · Model Card

Related articles