Mohamed Bouzertit

I’m a data annotator and AI evaluator with hands-on experience supporting large-scale language model training. I apply detailed guidelines to produce consistent outputs and audit results to keep data trustworthy.

In my Data Annotator role at RWS Train AI, I annotated and evaluated large volumes of text data using project instructions that evolve over time. I audited completed work to ensure consistency, accuracy, and alignment with quality standards, while identifying errors, edge cases, and unclear outputs that could impact model performance.

As an AI Evaluator at Outlier, I reviewed AI-generated responses for relevance, factual accuracy, and logical structure. I compared multiple outputs, selected the strongest using evaluation criteria, and flagged weak reasoning, hallucinations, and inconsistencies to improve future model behavior.

I’m comfortable in fast-paced remote environments with strict quality thresholds and deadlines. I focus on analytical thinking and careful decision-making—using structured scoring systems and clear written feedback to maintain reliable judgments across high-volume datasets.