Sobeir Ayami

Key Skills

Software

Appen

Lionbridge

Scale AI

Telus

Top Subject Matter

Medical and Scientific Content

General AI Safety & RLHF

Linguistic & Multilingual AI Training

Top Data Types

Text

Image

Video

Top Task Types

Data Collection

Bounding Box

Evaluation/Rating

Transcription

RLHF

Text Summarization

Question Answering

Text Generation

Classification

Polygon

Segmentation

Entity (NER) Classification

Object Detection

Red Teaming

Prompt + Response Writing (SFT)

Freelancer Overview

I am a medical student with extensive experience in AI training and data annotation, having worked on complex projects for platforms such as Outlier and Appen (Crowdgen). My expertise lies in RLHF methodology, where I evaluate and rank AI-generated responses based on accuracy, safety, and logical reasoning. I specialize in medical and scientific fact-checking, utilizing my clinical background to verify complex information and ensure high-quality training data is being provided. Beyond domain-specific knowledge, I am highly proficient in general AI-specific skills such as prompt engineering and navigating intricate annotation guidelines. I am a native German and Dari speaker, fluent in English, and experienced in multilingual content evaluation. My background in medicine, combined with a strong analytical mindset, allows me to provide precise, high-value feedback that helps fine-tune LLMs for specialized and general applications alike.

Labeling Experience

Remote AI Content Evaluator & Trainer

TextRLHF

As an AI Content Evaluator & Trainer at Outlier AI, I evaluated and ranked AI-generated responses with a focus on accuracy, safety, and reasoning. I applied RLHF principles to improve Large Language Models through targeted fine-tuning. Deep research and scientific verification ensured the creation of high-quality training data sets. • Conducted evaluation and ranking of language model outputs • Applied reinforcement learning from human feedback (RLHF) techniques • Verified complex medical and scientific information • Used proprietary tools for annotation and workflow management.

As an AI Content Evaluator & Trainer at Outlier AI, I evaluated and ranked AI-generated responses with a focus on accuracy, safety, and reasoning. I applied RLHF principles to improve Large Language Models through targeted fine-tuning. Deep research and scientific verification ensured the creation of high-quality training data sets. • Conducted evaluation and ranking of language model outputs • Applied reinforcement learning from human feedback (RLHF) techniques • Verified complex medical and scientific information • Used proprietary tools for annotation and workflow management.

2026 - Present

Remote Data Contributor / AI Specialist

AppenTextData Collection

At Appen (Crowdgen), I contributed to data collection and annotation projects focused on search relevance and model optimization. I Adhered to strict annotation and quality assurance guidelines while maintaining consistency. Managed workflow and tasks via various proprietary annotation platforms and tools. • Performed text data annotation for search and relevance tasks • Ensured high data quality through rigorous QA measures • Followed project-specific and proprietary annotation guidelines • Utilized internal platforms and quality-check software for workflow

At Appen (Crowdgen), I contributed to data collection and annotation projects focused on search relevance and model optimization. I Adhered to strict annotation and quality assurance guidelines while maintaining consistency. Managed workflow and tasks via various proprietary annotation platforms and tools. • Performed text data annotation for search and relevance tasks • Ensured high data quality through rigorous QA measures • Followed project-specific and proprietary annotation guidelines • Utilized internal platforms and quality-check software for workflow

2025 - Present

Education

Work History