John N.

Key Skills

Software

Appen

Clickworker

Data Annotation Tech

Remotasks

Top Subject Matter

Public Health & Epidemiology / Biomedical & Scientific content / YMYL safety content

Medical & Scientific text / Search relevance (SERP) / YMYL content safety

RLHF / LLM evaluation / Public health

Top Data Types

Text

Image

Document

Top Task Types

Entity (NER) Classification

RLHF

Segmentation

Classification

Object Detection

Transcription

Text Generation

Question Answering

Data Collection

Freelancer Overview

Domain Expert Evaluator — Health & Science Content (TELUS International AI / Lionbridge AI). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose, Appen, and DataAnnotation Tech. Education includes Bachelor of Science, Maseno University (2017). AI-training focus includes data types such as Text, Image, and Medical and labeling workflows including Evaluation, Rating, and Entity (NER) Classification.

Labeling Experience

Freelance AI Trainer & RLHF Specialist (DataAnnotation.tech / Outlier AI / Scale AI)

TextRLHF

Completed RLHF preference-ranking tasks for large language model fine-tuning with a focus on evidence-based judgments. Evaluated factual accuracy, logical coherence, and tone of LLM outputs across scientific, statistical, and public-health domains while flagging hallucinations. Authored prompt–response pairs for instruction fine-tuning and participated in calibration to resolve edge-case disagreements. • Completed 1,800+ RLHF preference-ranking tasks • Achieved >95% inter-annotator agreement against gold-standard labels • Delivered 10,000+ labeled data units across text, image, and structured formats for LLM/search-quality systems • Wrote prompt–response pairs covering data analysis, epidemiology, and research methodology

2024 - Present

AI Data Annotator — Text & Biomedical Track (Appen Medical & Scientific Content)

AppenTextEntity (NER) Classification

Labeled text for sentiment classification, intent detection, and named entity recognition using detailed schema taxonomies. Performed SERP relevance rating by evaluating helpfulness and other guideline-driven criteria for search-quality datasets. Identified systematic labeling drift within a batch and escalated it to QA leads to support guideline clarification and team consistency. • Delivered 3,500+ labeled text samples • Maintained quality scores above the platform threshold for three consecutive review periods • Independently flagged labeling drift in a 400-sample batch • Performed search result relevance rating using official guidelines

2023 - 2024

Medical Metadata & Ontology-Aligned Structured Labeling (Remotasks / Scale AI)

OtherClassification

Applied medical and ontology-aligned labels to structured and imaging-adjacent metadata to support downstream ML ingestion. Used schema-driven labeling and automated checks to keep batch accuracy high. Supported model training by producing consistent, audit-ready annotations for health-focused datasets. • Used ontology-aligned labeling for ML pipeline ingestion • Produced consistent structured labels for public-health survey data • Completed qualification training for vision labeling tasks • Maintained low error rates using automated verification

2023 - 2023

Image & Structured Data Annotator (Remotasks / Scale AI)

ImageSegmentation

Completed qualification training for 2D bounding box, polygon segmentation, and semantic classification for computer-vision datasets. Annotated medical imaging metadata and tabular public-health survey data using ontology-aligned labels for ML pipeline ingestion. Maintained high annotation throughput with low error rates validated by golden-task injection checks. • Completed Remotasks qualification training (2D bounding box, polygon segmentation, semantic classification) • Annotated medical imaging metadata and public-health survey tabular data • Maintained 200+ annotations per session • Achieved <2% error rate via automated golden-task checks

2023 - 2023

Domain Expert Evaluator — Health & Science Content (TELUS International AI / Lionbridge AI)

Don't discloseText

Performed search quality rating using YMYL guidelines, scoring query–result pairs for helpfulness, E-E-A-T, and content freshness. Ensured outputs met safety-critical health, medical, and scientific requirements by applying domain-specific standards during evaluation. Contributed to multilingual text classification by labeling English and Swahili samples for language identification and toxicity detection. • Rated 5,000+ query–result pairs • Selected for a Search Quality Rater specialist program after passing a 120-page YMYL guideline exam • Applied YMYL standards to health, medical, and scientific queries • Labeled English/Swahili samples for language identification and toxicity detection

2022 - 2023

Education

Work History