Sam Saleh

Key Skills

Software

iMerit

Scale AI

SuperAnnotate

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Medical Dicom

Text

Top Task Types

Evaluation Rating

Object Detection

Question Answering

RLHF

Translation Localization

Freelancer Overview

I am a medical doctor and health systems specialist with extensive experience in clinical data quality assurance, digital health innovation, and AI project oversight. My background includes managing rigorous data validation protocols for large-scale medical datasets, ensuring accuracy and reliability for decision-support tools in digital health and AI environments. I have also amasses extensive experience in Arabic-English translation, proofreading, localization, and AI training. I worked with Mercor ai and Scale ai on RLHF and Red-teaming. I am passionate about ensuring high-quality, ethically sourced training data to support advanced AI applications.

Labeling Experience

AI Red-Teamer

TextRLHF

RLHF project - Phase 1, 2, 3

RLHF project - Phase 1, 2, 3

2026 - Present

AI Evaluation Specialist (Medical & Multimodal)

Scale AITextEvaluation Rating

Evaluated large language model (LLM) outputs for clinical accuracy and safety in medical contexts. Applied comprehensive human factors and safety rubrics to assess vision-language model performance on medical images and structured tables. Identified hallucinations, incorrect medical advice, and ensured alignment with clinical best practices. • Conducted high-complexity model evaluations on diverse prompts. • Authored detailed, technical rationales to improve model safety. • Assessed multimodal outputs including text, images, and tabular data. • Supported iterative tuning and alignment for enhanced healthcare model reliability.

Evaluated large language model (LLM) outputs for clinical accuracy and safety in medical contexts. Applied comprehensive human factors and safety rubrics to assess vision-language model performance on medical images and structured tables. Identified hallucinations, incorrect medical advice, and ensured alignment with clinical best practices. • Conducted high-complexity model evaluations on diverse prompts. • Authored detailed, technical rationales to improve model safety. • Assessed multimodal outputs including text, images, and tabular data. • Supported iterative tuning and alignment for enhanced healthcare model reliability.

2025

Education

Work History