Arabic/English Bilingual AI Data Annotator & Evaluator

Role overview

This is a fully remote contractor position paid by the hour and organized around project needs. The work is flexible and may involve different AI language and data tasks depending on the assignment.

What you may work on

• Annotating data by labeling, categorizing, and organizing documents, tables, and other materials for AI training sets.
• Assessing AI-generated answers for correctness, reasoning quality, flow, and whether they fit the language and culture of the target audience.
• Creating high-quality prompts and sample model outputs to help train AI systems across a variety of subjects.
• Checking localized content for accuracy in terminology, tone, cultural nuance, and locale-specific details such as units, names, dates, and references.

Language and quality expectations

The role requires senior-level judgment in Arabic localization and translation, along with at least C1-level English proficiency, to keep terminology, tone, and cultural nuance accurate across different use cases.

About the company

CNTXT AI develops AI products and data solutions designed to make artificial intelligence accurate, safe, and globally relevant. Its work includes data services, custom AI solutions, and proprietary products, with strong specialization in Arabic-native and secure sovereign systems.

Profile requirements

• A bachelor’s degree or higher in Translation, Linguistics, Localization, Communications, or a closely related discipline.
• Native or near-native Arabic ability, including strong writing and editing skills in Modern Standard Arabic plus at least one dialect.
• At least C1-level English proficiency for reading and writing, especially when working with prompts, source material, and evaluation instructions.
• Strong editorial judgment around register, tone, punctuation, inclusivity, and cultural nuance, with a very high level of attention to detail.
• Ability to spot meaning shifts, ambiguity, localization mismatches, and subtle errors, and to explain fixes clearly in writing.
• Capability to verify localized content carefully using reliable sources and consistent reasoning, including units, references, names, and dates.
• Ability to detect reasoning flaws, methodological issues, and unclear explanations even when the language itself appears fluent.
• A dependable, self-managed working style with clear communication and responsiveness across time zones.

Preferred background

• Hands-on experience in localization or translation work.
• Exposure to MQM/LQA concepts such as severity, category, and root-cause analysis for consistent quality decisions.
• Comfort using CAT tools and QA workflows, including translation memory, termbases, and automated quality checks.
• Prior work in AI data training, annotation, or evaluation.

Additional information

The role is project-based and may shift across annotation, evaluation, content generation, and localization QA depending on business needs. No fixed on-site schedule is required, and collaboration may occur across time zones.