AI Dataset Curator: Build Clean, Ethical Training Data

Odixcity Consulting; La Réunion; Sur place; EUR 42 000 - 65 000; Plein temps

Résumé du poste

Odixcity Consulting is seeking a Dataset Curator who designs, maintains, and optimizes high-quality AI/ML datasets. You will work with data scientists, AI trainers, and engineers to gather, clean, validate, and annotate data across domains, ensuring data supports robust model training and evaluation.

Responsibilities include bias detection, metadata management, and ethical data sourcing, with scalable workflows for large-scale datasets used in model assessment and deployment.

Qualifications

• Bachelor’s or Master’s degree in data-related field.
• Proficiency in Excel, Google Sheets, SQL, and/or Python.
• Knowledge of cleaning, normalization, and transformation of data.
• Familiarity with data annotation tools.
• Understanding of structured to unstructured datasets.
• 3-5 years of experience in dataset curation or data management.
• Experience with large-scale datasets for AI/ML.

Responsabilités

• Curate, collect, and structure datasets for AI/ML training.
• Validate dataset accuracy, completeness, and consistency.
• Annotate and label datasets according to guidelines.
• Identify and correct data inconsistencies, duplicates, and anomalies.
• Maintain metadata and documentation for datasets.
• Collaborate with AI trainers and data engineers to define requirements.
• Ensure datasets are ethically sourced and free from biases.
• Continuously monitor dataset quality and propose improvements.

Connaissances

Dataset curation

Data cleaning

Data annotation

Metadata management

Bias detection

Python

SQL

Excel/Sheets

Formation

Bachelor’s or Master’s degree in a relevant field

Outils

Excel

SQL

Python

Git / Version control

Description du poste

Odixcity Consulting is seeking a Dataset Curator who designs, maintains, and optimizes high-quality AI/ML datasets. You will work with data scientists, AI trainers, and engineers to gather, clean, validate, and annotate data across domains, ensuring data supports robust model training and evaluation.

Responsibilities include bias detection, metadata management, and ethical data sourcing, with scalable workflows for large-scale datasets used in model assessment and deployment.