For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Yusuf Elgharbawy

Yusuf Elgharbawy

AI Data Specialist & Multimodal Trainer

Egypt flagDamietta, Egypt
$10.00/hrExpertData Annotation TechOther

Key Skills

Software

Data Annotation TechData Annotation Tech
Other

Top Subject Matter

Multimodal LLM Personalization and Fine-tuning
Text-to-Image and Video-to-Image Generative AI
Video Genre

Top Data Types

ImageImage
TextText
VideoVideo
AudioAudio

Top Task Types

Evaluation Rating
Fine Tuning
RLHF
Text Generation
Prompt Response Writing SFT
Classification
Transcription

Freelancer Overview

AI Data Specialist & Multimodal Trainer. Core strengths include Internal and Proprietary Tooling. AI-training focus includes data types such as Text, Image, and Video and labeling workflows including Prompt + Response Writing (SFT), Evaluation, and Rating.

ExpertArabicEnglish

Labeling Experience

AI Data Specialist & Multimodal Trainer

AudioTranscription
I performed precise bilingual (English/Arabic) audio transcription and multi-dimensional quality assessments for speech datasets. Each segment underwent linguistic and phonetic analysis for accuracy, clarity, and adherence to study objectives. This work was essential for training and validating speech recognition and understanding algorithms. • Transcribed medical, conversational, and technical speech data. • Conducted 6-dimensional qualitative reviews for each audio file. • Collaborated in cross-disciplinary teams for error analysis and correction. • Delivered consistently high approval rates on transcription quality.

I performed precise bilingual (English/Arabic) audio transcription and multi-dimensional quality assessments for speech datasets. Each segment underwent linguistic and phonetic analysis for accuracy, clarity, and adherence to study objectives. This work was essential for training and validating speech recognition and understanding algorithms. • Transcribed medical, conversational, and technical speech data. • Conducted 6-dimensional qualitative reviews for each audio file. • Collaborated in cross-disciplinary teams for error analysis and correction. • Delivered consistently high approval rates on transcription quality.

2025 - Present

AI Data Specialist & Multimodal Trainer

VideoClassification
I annotated and classified complex video datasets, focusing on genre, geolocation, and prompt-driven content verification. Each annotation followed detailed internal rubrics ensuring consistency and accuracy for supervised learning workflows. The work contributed to the evaluation and improvement of video model outputs for genre and topographic fidelity. • Applied prompt-based criteria to segment and verify relevant video frames. • Annotated video data for geographic and thematic categorization. • Worked with diverse content from educational to entertainment domains. • Followed multi-step QA review cycles for high-label integrity.

I annotated and classified complex video datasets, focusing on genre, geolocation, and prompt-driven content verification. Each annotation followed detailed internal rubrics ensuring consistency and accuracy for supervised learning workflows. The work contributed to the evaluation and improvement of video model outputs for genre and topographic fidelity. • Applied prompt-based criteria to segment and verify relevant video frames. • Annotated video data for geographic and thematic categorization. • Worked with diverse content from educational to entertainment domains. • Followed multi-step QA review cycles for high-label integrity.

2025 - Present

AI Data Specialist & Multimodal Trainer

Image
I conducted thorough multimodal model evaluation with a focus on comparing text-to-image and video-to-image outputs at multiple stages of development. Each evaluation emphasized photorealism, visual consistency, and adherence to data-specific metadata for quality assurance. Findings were documented and communicated directly to development and research teams for further improvement. • Analyzed and rated visual content accuracy and style. • Detected inconsistencies in AI-generated imagery and flagged errors. • Cross-referenced outputs with reference data and prompts for completeness. • Ensured compliance with privacy, ethical, and internal review standards.

I conducted thorough multimodal model evaluation with a focus on comparing text-to-image and video-to-image outputs at multiple stages of development. Each evaluation emphasized photorealism, visual consistency, and adherence to data-specific metadata for quality assurance. Findings were documented and communicated directly to development and research teams for further improvement. • Analyzed and rated visual content accuracy and style. • Detected inconsistencies in AI-generated imagery and flagged errors. • Cross-referenced outputs with reference data and prompts for completeness. • Ensured compliance with privacy, ethical, and internal review standards.

2025 - Present

AI Data Specialist & Multimodal Trainer

TextPrompt Response Writing SFT
I authored and engineered sophisticated multi-paragraph prompts for LLM personalization and fine-tuning across research and production environments. The role demanded high-level understanding of prompt syntax, structure, and intended model behavior to ensure optimal training outcomes. Rigorous review and iteration of prompts were conducted based on feedback and evolving requirements throughout the training lifecycle. • Developed prompts addressing specialized technical and mathematical datasets. • Ensured clarity, diversity, and depth in input scenarios for model robustness. • Collaborated with multilingual teams to guarantee cross-lingual applicability. • Maintained exhaustive documentation for audit and reproducibility purposes.

I authored and engineered sophisticated multi-paragraph prompts for LLM personalization and fine-tuning across research and production environments. The role demanded high-level understanding of prompt syntax, structure, and intended model behavior to ensure optimal training outcomes. Rigorous review and iteration of prompts were conducted based on feedback and evolving requirements throughout the training lifecycle. • Developed prompts addressing specialized technical and mathematical datasets. • Ensured clarity, diversity, and depth in input scenarios for model robustness. • Collaborated with multilingual teams to guarantee cross-lingual applicability. • Maintained exhaustive documentation for audit and reproducibility purposes.

2025 - Present

Text Annotation & Model Output Evaluation (LLM Projects)

Don T DiscloseTextRLHFFine Tuning
Worked on text annotation and AI model evaluation projects across multiple platforms. Responsibilities included: • Comparing receipts from different retail stores to ensure data consistency and structured alignment • Rubrics alignment and evaluation based on predefined quality criteria • Comparing outputs from different AI models and selecting the best response according to accuracy, clarity, and instruction adherence • Writing and refining Arabic prompts for LLM training and evaluation • Identifying logical inconsistencies, factual errors, and formatting issues in generated responses Maintained high accuracy standards, followed detailed guidelines, and ensured consistency across large batches of annotated data. Demonstrated strong analytical thinking and attention to detail while working independently in remote environments.

Worked on text annotation and AI model evaluation projects across multiple platforms. Responsibilities included: • Comparing receipts from different retail stores to ensure data consistency and structured alignment • Rubrics alignment and evaluation based on predefined quality criteria • Comparing outputs from different AI models and selecting the best response according to accuracy, clarity, and instruction adherence • Writing and refining Arabic prompts for LLM training and evaluation • Identifying logical inconsistencies, factual errors, and formatting issues in generated responses Maintained high accuracy standards, followed detailed guidelines, and ensured consistency across large batches of annotated data. Demonstrated strong analytical thinking and attention to detail while working independently in remote environments.

2024

Education

A

Al-Azhar University

Bachelor of Medicine, Bachelor of Surgery, Medicine

Bachelor of Medicine, Bachelor of Surgery
2023 - 2026

Work History

F

Freelance / Self-Employed

Freelance AI Data Annotator

Damietta
2024 - Present
L

Local Community Pharmacy

Pharmacy Assistant

Damietta
2024 - 2024