Remote Armenian AI Linguist & Data Curator

Confidential; United States; Remote; USD 60,000 - 100,000; Full time

Job summary

Confidential is seeking experienced Armenian Linguistic Experts for a 100% remote contract project in the United States. You will evaluate LLM Armenian outputs, edit content for linguistic and cultural accuracy, and annotate texts to improve model training.

Strong English proficiency and native or near-native Armenian fluency are required. The role focuses on data curation, quality assessment, and collaboration with AI researchers to close language gaps and enhance model performance.

Qualifications

• Native or near-native fluency in Armenian with excellent grammar and style.
• Professional experience in translation, editing, linguistic annotation, or NLP data work in Armenian.
• Strong understanding of Armenian nuances, variations, and context-appropriate usage.
• Professional fluency in English.

Responsibilities

• Evaluate the quality and accuracy of LLM-generated Armenian text across use cases (translation, summarization, dialogue, grammar).
• Edit, refine, or rewrite Armenian content to meet linguistic and cultural standards for Armenian audiences.
• Annotate Armenian text for grammatical, syntactic, and semantic features.
• Identify model errors and language gaps, and collaborate with AI researchers to improve LLM performance.
• Provide detailed linguistic feedback and analysis to guide ongoing model training efforts.

Skills

Armenian fluency

English proficiency

Translation/editing

Linguistic annotation

NLP data work

Tools

Annotation tools

Job description

Confidential is seeking experienced Armenian Linguistic Experts for a 100% remote contract project in the United States. You will evaluate LLM Armenian outputs, edit content for linguistic and cultural accuracy, and annotate texts to improve model training.

Strong English proficiency and native or near-native Armenian fluency are required. The role focuses on data curation, quality assessment, and collaboration with AI researchers to close language gaps and enhance model performance.