Image Data Annotator for AI Vision | Impact & Precision

Harmattan AI; Paris; Sur place; EUR 24 000 - 30 000; Plein temps

Résumé du poste

Harmattan AI is seeking a Data Annotator within the Data Engineering team to label and annotate image datasets for autonomous defense systems. You will collaborate with Computer Vision, Data Architecture, and ML teams to ensure data quality for real-world deployment.

Applicants should be meticulous, capable of long, repetitive tasks, and fluent in French, with English as a plus. This role supports a structured, guideline-driven workspace and may involve on-site data collection in some campaigns.

Qualifications

• Must have a baccalaureat or equivalent.
• Meticulous attention to detail and accuracy across long tasks.
• Fluent in French; English a plus.
• Comfortable with digital tools and following strict project guidelines.
• Experience in data labeling is a plus.

Responsabilités

• Analyze, label, and annotate diverse image datasets to support AI model training and validation.
• Follow detailed annotation guidelines precisely, maintaining consistency across large batches of work.
• Review AI model outputs for accuracy and contextual relevance, flagging discrepancies.
• Perform annotation peer-review with other members of the Annotation team to ensure quality.
• Communicate findings and edge cases to the AI and Data Engineering teams.
• Occasionally support field test campaigns to help with on-site data collection.

Connaissances

Attention to detail

Written and verbal communication

French fluency

Data labeling

Guideline compliance

Formation

Baccalauréat

Description du poste

Harmattan AI is seeking a Data Annotator within the Data Engineering team to label and annotate image datasets for autonomous defense systems. You will collaborate with Computer Vision, Data Architecture, and ML teams to ensure data quality for real-world deployment.

Applicants should be meticulous, capable of long, repetitive tasks, and fluent in French, with English as a plus. This role supports a structured, guideline-driven workspace and may involve on-site data collection in some campaigns.