Mahmoud Mahmoud

Key Skills

Software

Other

Labelbox

CVAT

Top Subject Matter

No subject matter listed

Top Data Types

Text

Medical Dicom

Video

Image

Top Task Types

Bounding Box

Classification

Diagnosis

Entity Ner Classification

Evaluation Rating

Object Detection

Prompt Response Writing SFT

Question Answering

RLHF

Text Generation

Text Summarization

Tracking

Transcription

Freelancer Overview

I bring over eight years of experience in clinical research, medical affairs, and epidemiology, with a strong focus on large-scale data collection, annotation, and analysis in medical and healthcare domains. My background includes leading cohort studies involving over 50,000 patients, supporting real-world evidence (RWE) generation, and ensuring strict compliance with international data standards such as ICH-GCP. I have collaborated cross-functionally with clinical, regulatory, and technical teams to deliver high-quality, accurate datasets for research and regulatory submissions. My expertise spans medical imaging, cohort data, and device performance validation, and I am skilled in scientific writing, statistical analysis, and database management. I am passionate about leveraging my medical and analytical skills to support AI development through precise data labeling, annotation, and quality assurance, especially in projects involving medical data, imaging, and healthcare AI.

Labeling Experience

Generalists

OtherImageBounding BoxObject Detection

High quality annotation that includes image, text and video labeling.

High quality annotation that includes image, text and video labeling.

2024 - 2025

Video Annotation Specialist – Computer Vision

CVATVideoBounding BoxClassification

Annotated video datasets for computer vision model training, performing frame-by-frame labeling and object tracking across video sequences. Applied bounding boxes and event tags to identify objects, actions, and temporal changes within videos. Followed detailed annotation guidelines to ensure consistency across frames and maintain high labeling accuracy. Conducted quality checks to verify correct object continuity and alignment across time, contributing to improved video model performance.

Annotated video datasets for computer vision model training, performing frame-by-frame labeling and object tracking across video sequences. Applied bounding boxes and event tags to identify objects, actions, and temporal changes within videos. Followed detailed annotation guidelines to ensure consistency across frames and maintain high labeling accuracy. Conducted quality checks to verify correct object continuity and alignment across time, contributing to improved video model performance.

2025

LLM Data Annotation & Alignment Specialist

OtherTextClassificationText Generation

Worked on large-scale LLM training and alignment projects using Outlier , Handshakeai , Multimango and Aligner platforms. Responsibilities included evaluating AI-generated responses, ranking outputs based on quality and relevance, and ensuring compliance with detailed annotation and safety guidelines. Performed instruction-following assessments, content accuracy checks, and preference labeling to improve model behavior and response quality. Maintained high annotation accuracy while meeting productivity targets and adhering to strict quality assurance standards.

Worked on large-scale LLM training and alignment projects using Outlier , Handshakeai , Multimango and Aligner platforms. Responsibilities included evaluating AI-generated responses, ranking outputs based on quality and relevance, and ensuring compliance with detailed annotation and safety guidelines. Performed instruction-following assessments, content accuracy checks, and preference labeling to improve model behavior and response quality. Maintained high annotation accuracy while meeting productivity targets and adhering to strict quality assurance standards.

2025

AI Response Quality Analyst (RLHF)

LabelboxMedical DicomEntity Ner ClassificationQuestion Answering

Reviewed and compared multiple AI-generated responses to determine optimal outputs based on accuracy, helpfulness, tone, and safety guidelines. Supported reinforcement learning from human feedback (RLHF) workflows by providing consistent and high-quality preference labels.

Reviewed and compared multiple AI-generated responses to determine optimal outputs based on accuracy, helpfulness, tone, and safety guidelines. Supported reinforcement learning from human feedback (RLHF) workflows by providing consistent and high-quality preference labels.

2025 - 2025

Education

Work History