Senior AI Trainer - RLHF, Data Labeling & Evaluation, Remote

Rex.zone; United States; Sur place; USD 41 328 - 68 880; Plein temps

Résumé du poste

Rex.zone is looking for an experienced AI Trainer for a remote, full-time position focusing on improving AI systems. You'll be tasked with executing RLHF, data labeling, prompt evaluation, and QA evaluation to enhance model quality. Candidates should have mid-senior experience in data workflows and strong communication skills. Compensation is competitive at $30–$50/hr, with potential tasks across NLP, computer vision, and content safety labeling workflows.

Qualifications

• Mid-senior experience in data labeling, evaluation, or applied AI workflows.
• Strong written communication for rubric-based judging and rationale writing.
• Ability to follow detailed guidelines with high inter-annotator consistency.
• Familiarity with RLHF concepts, model evaluation, and prompt evaluation.
• Comfort with spreadsheets and QA workflows.

Responsabilités

• Execute RLHF and preference ranking tasks on model outputs.
• Perform prompt evaluation and response grading using strict rubrics.
• Label and verify training data for NLP and computer vision annotation.
• Run QA evaluation to catch ambiguity, leakage, and policy violations.

Connaissances

Data labeling

Model evaluation

Rubric-based judging

Communication skills

Consistency in annotation

Spreadsheets

QA workflows

Description du poste

Rex.zone is looking for an experienced AI Trainer for a remote, full-time position focusing on improving AI systems. You'll be tasked with executing RLHF, data labeling, prompt evaluation, and QA evaluation to enhance model quality. Candidates should have mid-senior experience in data workflows and strong communication skills. Compensation is competitive at $30–$50/hr, with potential tasks across NLP, computer vision, and content safety labeling workflows.