French AI Response Evaluator & Annotator

Talanto; France; Hybride; EUR 24 000 - 28 000; Plein temps

Avantages offerts par ce poste

Medical, dental, and vision coverage

Flexible Spending Account (FSA)

Parental leave

Professional growth and development

Résumé du poste

Talanto is seeking an AI Response Labeler / Annotator with deep French expertise to evaluate AI-generated responses, focusing on language, tone, and cultural context for France. The role involves side-by-side comparisons, assessing accuracy and usefulness across English and French content, with structured guidelines and high-volume tasks.

You will work in a highly structured environment, applying annotation rubrics, while maintaining quality and productivity, across diverse scenarios and data

Qualifications

• Native-level or professional fluency in French.
• Strong English reading comprehension and ability to understand guidelines written in English.
• Strong analytical and critical-thinking skills beyond language evaluation.
• Excellent written communication and ability to explain decisions clearly.

Responsabilités

• Perform side-by-side comparisons of AI-generated responses and determine which is stronger.
• Evaluate responses for factual accuracy, relevance, completeness, clarity, and tone.
• Assess content in English, French, or both depending on scenario.
• Evaluate a broad range of content including web results, file-based tasks, and multi-turn conversations.
• Apply detailed annotation guidelines consistently and independently.

Connaissances

French fluency

English fluency

Analytical thinking

Attention to detail

Independent work

Description du poste

Talanto is seeking an AI Response Labeler / Annotator with deep French expertise to evaluate AI-generated responses, focusing on language, tone, and cultural context for France. The role involves side-by-side comparisons, assessing accuracy and usefulness across English and French content, with structured guidelines and high-volume tasks.

You will work in a highly structured environment, applying annotation rubrics, while maintaining quality and productivity, across diverse scenarios and data