Salah Edeen Ali

Key Skills

Software

OneForma

Appen

Top Subject Matter

No subject matter listed

Top Data Types

Text

Top Label Types

RLHF

Translation Localization

Entity Ner Classification

Freelancer Overview

I possess practical experience in data annotation and labeling, supported by a specialized academic background from Asyut University in Arabic and English education. Unlike generic annotators, I approach data labeling with the rigorous eye of a linguist. My experience involves evaluating complex model outputs, identifying subtle syntactic errors, and ensuring semantic consistency across datasets. I am proficient in strict error identification protocols and quality evaluation, ensuring that training data meets the highest standards of grammatical integrity and accuracy. "What sets me apart is my 10-year track record as a professional translator and proofreader, which allows me to bring 'Subject Matter Expert' quality to data labeling. I specialize in the nuance of language—specifically the distinction between Modern Standard Arabic and Egyptian dialects. I leverage this cultural intelligence to identify non-native phrasing, correct hallucinations, and ensure cultural precision in AI responses. This combination of technical labeling experience and deep linguistic expertise enables me to deliver high-fidelity data that truly enhances model performance.

Labeling Experience

Arabic Machine Translation Post-Editing & Localization

AppenTextTranslation Localization

Lead linguist for a high-volume Machine Translation Post-Editing (MTPE) project aimed at refining an English-to-Arabic translation engine. Key tasks included: Semantic Correction: Reviewed machine-generated Arabic text against English source material to correct mistranslations and omissions. Syntax Validation: Fixed grammatical structures to ensure natural flow and adherence to formal Arabic rules. Cultural Localization: Adapted content to fit Egyptian cultural norms, ensuring tone and register were appropriate for the target audience. Project Size: Processed over [Insert Number, e.g., 50,000] words with a focus on high-fidelity linguistic data.

Lead linguist for a high-volume Machine Translation Post-Editing (MTPE) project aimed at refining an English-to-Arabic translation engine. Key tasks included: Semantic Correction: Reviewed machine-generated Arabic text against English source material to correct mistranslations and omissions. Syntax Validation: Fixed grammatical structures to ensure natural flow and adherence to formal Arabic rules. Cultural Localization: Adapted content to fit Egyptian cultural norms, ensuring tone and register were appropriate for the target audience. Project Size: Processed over [Insert Number, e.g., 50,000] words with a focus on high-fidelity linguistic data.

2025 - 2025

Arabic Entity Extraction (NER) & Classification

OneformaTextEntity Ner Classification

Performed detailed data annotation for an NLP model focused on Entity Recognition and Sentiment Analysis. Key tasks included: Named Entity Recognition (NER): Accurately tagged entities such as Persons, Locations, Organizations, and Dates within unstructured Arabic text. Text Classification: Categorized user-generated content based on sentiment (Positive, Negative, Neutral) and intent. Ambiguity Resolution: Used linguistic expertise to resolve context-dependent ambiguities in the text. Quality Measures: Maintained consistency across large datasets to ensure reliable Ground Truth data for model training.

Performed detailed data annotation for an NLP model focused on Entity Recognition and Sentiment Analysis. Key tasks included: Named Entity Recognition (NER): Accurately tagged entities such as Persons, Locations, Organizations, and Dates within unstructured Arabic text. Text Classification: Categorized user-generated content based on sentiment (Positive, Negative, Neutral) and intent. Ambiguity Resolution: Used linguistic expertise to resolve context-dependent ambiguities in the text. Quality Measures: Maintained consistency across large datasets to ensure reliable Ground Truth data for model training.

2024 - 2024

Advanced Arabic LLM Training & RLHF Evaluation

OneformaTextRLHF

Served as a Linguistic Subject Matter Expert (SME) for a large-scale Generative AI project. The scope involved evaluating and ranking model responses based on Helpfulness, Honesty, and Harmlessness (HHH) criteria.

Served as a Linguistic Subject Matter Expert (SME) for a large-scale Generative AI project. The scope involved evaluating and ranking model responses based on Helpfulness, Honesty, and Harmlessness (HHH) criteria.

2023 - 2023

Education

Work History