Remote Data Entry Specialist for AI Benchmarking

RemoteJobsOne; Paris; À distance; EUR 24 000 - 43 000; Temps partiel

Résumé du poste

RemoteJobsOne is seeking a Data Entry Keyer for a fully remote contractor role based in France. You will design and curate expert-level data-entry tasks to train AI benchmarks, including handling real-world data with errors and variations.

Your work will shape how models learn and perform, with emphasis on accuracy and audit-ready outputs. Applicants should have strong data-entry/QA experience, excellent English writing, and a meticulous, process-driven approach to creating datasets and grading

Qualifications

• Experience in data entry, QA, or data validation in regulated environments.
• Proven ability to document high accuracy with measurable error rates.
• Strong skills in identifying data inconsistencies and formatting irregularities across file types.
• Proficient written English for drafting evaluation tasks and rubrics.
• Exceptional attention to detail and a process-driven work style.
• Experience contributing to high-stakes projects requiring data integrity.

Responsabilités

• Design and produce evaluation tasks simulating real-world data entry challenges.
• Create large datasets including CSVs, PDFs and spreadsheets with realistic errors.
• Define final states and reconciliation rules for AI training tasks.
• Draft detailed grading rubrics (35+ criteria) for AI outputs.

Connaissances

Data entry

Quality assurance

Data validation

Written English

Attention to detail

Remote collaboration

Outils

Spreadsheet software

CSV/PDF editors

Description du poste

RemoteJobsOne is seeking a Data Entry Keyer for a fully remote contractor role based in France. You will design and curate expert-level data-entry tasks to train AI benchmarks, including handling real-world data with errors and variations.

Your work will shape how models learn and perform, with emphasis on accuracy and audit-ready outputs. Applicants should have strong data-entry/QA experience, excellent English writing, and a meticulous, process-driven approach to creating datasets and grading