Elizabeth Z.

Key Skills

Software

Scale AI

Top Subject Matter

Clinical/medical domain AI evaluation and RLHF feedback

English language comprehension and AI response evaluation

Legal Services & Contract Review

Top Data Types

Text

Document

Top Task Types

Fine-tuning

Question Answering

Text Generation

Classification

Red Teaming

Text Summarization

Freelancer Overview

AI Data Evaluation Contributor, DataAnnotation.tech (Remote). Brings 13+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include DataAnnotation.tech and Scale AI. Education includes Bachelor of Science in Nursing, Eastern Michigan University (2019) and Bachelor of Science in Biology, University of Michigan (2013). AI-training focus includes data types such as Text and labeling workflows including Evaluation and Rating.

Labeling Experience

AI Model Output Evaluator, Outlier AI / Scale AI (Remote)

Scale AIText

Screen and qualify as a generalist evaluator for English language comprehension, composition, and advanced reading analysis within AI response assessment tasks. Evaluate AI-generated responses for quality, accuracy, and instruction-following using structured assessment frameworks. Provide ratings to support model evaluation and downstream quality improvements. • Assess instruction-following and response quality • Judge accuracy and reading/comprehension alignment • Apply structured assessment frameworks for consistency • Support dataset/model quality evaluation through ratings

2026 - Present

AI Data Evaluation Contributor, DataAnnotation.tech (Remote)

Text

Evaluate AI model outputs for clinical accuracy, logical reasoning, and adherence to prompt instructions using structured evaluation rubrics. Identify hallucinations, factual errors, and unsafe outputs in AI-generated clinical content with a focus on medical-domain correctness. Provide detailed failure summaries and qualitative feedback to support reinforcement learning from human feedback workflows. • Grade model performance across rubric-based assessments • Detect unsafe outputs and reasoning errors • Produce structured qualitative failure summaries • Perform comparative ranking for helpfulness, safety, and factual accuracy

2025 - Present

Education

Work History