Abdulrahman H.

Key Skills

Software

Micro1

Data Annotation Tech

SuperAnnotate

Top Subject Matter

Micro1 - Video Reviewer

SuperAnnotate -Text to Image

Top Data Types

Image

Video

Text

Top Task Types

Evaluation/Rating

Text Generation

Classification

Data Collection

Prompt + Response Writing (SFT)

Freelancer Overview

I have experience working with data annotation and AI training tasks that require strong attention to detail, consistency, and adherence to project guidelines. My work has involved labeling and categorizing text, reviewing data for quality and accuracy, identifying patterns, and ensuring datasets meet the standards required for machine learning model development. I am comfortable following detailed instructions, maintaining high accuracy rates, and meeting deadlines while handling large volumes of data. What sets me apart is my analytical mindset, ability to learn new annotation frameworks quickly, and commitment to delivering reliable results. I have strong written communication skills, excellent judgment when evaluating content, and experience using digital tools and online platforms efficiently. I am highly organized, dependable, and able to work independently while maintaining the quality standards necessary for AI training and data labeling projects.

Labeling Experience

Video annotator

VideoText Generation

On Micro 1, I handled video annotation for hand object interaction data used in action recognition models. I reviewed and refined model generated labels to ensure accuracy in hand movements, object interactions, and timing. My work focused on aligning action boundaries to the exact start and end frames, ensuring each caption described a single atomic action, and verifying correct left vs. right hand usage in every segment. I maintained labeling quality by strictly following the project’s annotation rubric and structured review pipeline. I flagged ambiguous or low quality labels for rework, collaborated with team leads to resolve edge cases, and provided feedback to improve model pre labels. This ensured our data met the precision standards required for downstream model training and reduced work rates. Project size was 500 experts.

2026 - Present

Prompt evaluator

TextPrompt + Response Writing (SFT)

At Meta, I work on prompt engineering and response evaluation to improve AI model behavior. I design prompts that test specific capabilities like reasoning, instruction following, and safety, often targeting edge cases or nuanced user requests. I then evaluate model outputs across multiple dimensions including accuracy, helpfulness, tone, and I adhere to Meta’s evaluation rubrics and quality guidelines to ensure consistency and objectivity. This includes doing side by side comparisons, scoring responses against defined criteria, and providing detailed rationales for each rating. I also flag failure patterns and ambiguous cases to help refine guidelines, which directly improves model training and evaluation standards.

2023 - 2025

Education

Work History