Jamal Maàdini

Key Skills

Software

Appen

CrowdSource

Labelbox

Mindrift

Toloka

Telus

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Video

Top Task Types

Data Collection

Evaluation Rating

Object Detection

Prompt Response Writing SFT

Text Generation

Freelancer Overview

I have extensive experience in data labeling and AI training across multiple domains, including advertising, language evaluation, and multimodal content analysis. My work has involved annotating and validating datasets for text, image, and video tasks, ensuring high-quality input for machine learning models. I have specialized in multilingual content evaluation (French, English, Arabic, Moroccan Darija), focusing on translation accuracy, semantic consistency, cultural adaptation, and sensitivity review. In addition, I have contributed to projects such as ad relevance and compliance evaluation, LLM output review, and lip-sync/video alignment assessments, which required both linguistic expertise and attention to detail. I am skilled at applying complex guidelines, handling ambiguity, and maintaining consistency at scale—qualities that make me well-suited for high-impact AI training and data quality projects.

Labeling Experience

English-Arabic Translation Data Annotator / Content Reviewer

AppenAudioEvaluation RatingData Collection

The Appen project focused on improving AI translation systems and content moderation tools. Tasks included reviewing English to Arabic translations for accuracy, fluency, and cultural appropriateness, as well as identifying and labeling harmful or inappropriate content. The project was large-scale and required strict adherence to guidelines, quality checks, and accuracy standards to ensure reliable training data for AI models.

The Appen project focused on improving AI translation systems and content moderation tools. Tasks included reviewing English to Arabic translations for accuracy, fluency, and cultural appropriateness, as well as identifying and labeling harmful or inappropriate content. The project was large-scale and required strict adherence to guidelines, quality checks, and accuracy standards to ensure reliable training data for AI models.

2024

Ads Assessor

TelusTextEvaluation Rating

The TELUS International rating projects focus on improving AI and search performance by providing large-scale, high-quality human judgments on ads, search results, and content. Tasks include evaluating ad relevance, landing page quality, multilingual AI outputs, multimodal data (text, image, video), and flagging sensitive content. These projects involve thousands of raters worldwide and rely on strict guideline compliance, regular audits, and accuracy thresholds to ensure consistent, reliable training data for AI systems.

The TELUS International rating projects focus on improving AI and search performance by providing large-scale, high-quality human judgments on ads, search results, and content. Tasks include evaluating ad relevance, landing page quality, multilingual AI outputs, multimodal data (text, image, video), and flagging sensitive content. These projects involve thousands of raters worldwide and rely on strict guideline compliance, regular audits, and accuracy thresholds to ensure consistent, reliable training data for AI systems.

2024

Multilingual Data Annotator

MindriftImageText GenerationAction Recognition

The Mindrift project focused on improving AI performance through multilingual data annotation and content evaluation. Tasks included describing images, evaluating translations, and reviewing text quality in Arabic, Moroccan Darija, French, and English. The project involved large-scale data labeling with strict adherence to guidelines, quality checks, and accuracy measures to ensure reliable training data for AI systems.

The Mindrift project focused on improving AI performance through multilingual data annotation and content evaluation. Tasks included describing images, evaluating translations, and reviewing text quality in Arabic, Moroccan Darija, French, and English. The project involved large-scale data labeling with strict adherence to guidelines, quality checks, and accuracy measures to ensure reliable training data for AI systems.

2024 - 2025

Education

Work History