Mohamed Taha Aniba

Key Skills

Software

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Text

Top Task Types

Computer Programming Coding

Evaluation Rating

Prompt Response Writing SFT

RLHF

Translation Localization

Freelancer Overview

I have hands-on experience in AI training and data labeling, gained through impactful projects at Teleperformance and Outlier. At Teleperformance, I contributed to advanced AI systems, particularly on the Meta Ray-Ban and Meta Language Model projects, where I performed complex annotation, quality evaluation, and linguistic data categorization. These tasks required high attention to detail, language sensitivity (Arabic, French, and English), and adherence to strict quality and confidentiality guidelines. At Outlier, I worked on various AI training tasks such as prompt ranking, rewriting, transcription evaluation, and multilingual data annotation, following rigorous guidelines to improve large language model performance. My ability to quickly adapt to new guidelines, work with nuanced instructions, and deliver consistent, high-quality outputs across different projects sets me apart as a reliable and skilled AI training contributor.

Labeling Experience

Multilingual AI Prompt Evaluation & Transcription Labeling

Scale AITextText GenerationTranslation Localization

This project focused on improving large language model output across multiple languages (Arabic, English, French). Tasks included ranking AI-generated responses, rewriting for clarity or correctness, evaluating factual accuracy, and reviewing transcripts with detailed error annotation (including filler words, hesitations, false starts). I followed strict guidelines to ensure consistency, relevance, and linguistic accuracy. The project required strong language judgment, adherence to nuanced instructions, and rapid adaptation to evolving quality standards.

This project focused on improving large language model output across multiple languages (Arabic, English, French). Tasks included ranking AI-generated responses, rewriting for clarity or correctness, evaluating factual accuracy, and reviewing transcripts with detailed error annotation (including filler words, hesitations, false starts). I followed strict guidelines to ensure consistency, relevance, and linguistic accuracy. The project required strong language judgment, adherence to nuanced instructions, and rapid adaptation to evolving quality standards.

2024

Education

Work History