Global AI lab
Under NDAArabic evaluation desk
Native-dialect eval suites, preference data, and red-team probes for MENA model releases.
Portfolio
We maintain a vetted bench of Arabic reviewers, RLHF specialists, and domain experts — deployed on scoped client work under NDA.
Client names and metrics are shared under NDA during scoping. Below are representative engagement types.
Global AI lab
Under NDANative-dialect eval suites, preference data, and red-team probes for MENA model releases.
Financial services
Under NDARetrieval harness, Arabic chunking review, and board-ready accuracy report with remediation plan.
Data company
Under NDACurated corpora, rubric design, reviewer QA, and export to client training infrastructure.
Bench
Next step
One form for everyone — businesses, AI labs, and partners. A human reads it and replies within one business day.