Her_zhee Z.

Key Skills

Software

Scale AI

Appen

Other

Top Subject Matter

Reinforcement learning from human feedback (RLHF) and LLM alignment quality training

ML data annotation and taxonomy management

Conversational AI dataset creation and prompt-response training data

Top Data Types

Text

Image

Document

Top Task Types

RLHF

Classification

Prompt + Response Writing (SFT)

Freelancer Overview

Senior AI Trainer & Quality Lead, Outlier AI (Scale AI Partner). Brings 8+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Scale AI, Appen, and Other. AI-training focus includes data types such as Computer Code, Programming, and Text and labeling workflows including RLHF, Classification, and Prompt + Response Writing (SFT).

Labeling Experience

Senior AI Trainer & Quality Lead, Outlier AI (Scale AI Partner)

Scale AIRLHF

Provided RLHF feedback and quality-focused annotations to improve model behavior across reasoning, coding, and math domains. Ensured training output met high acceptance standards through rubric-based evaluation and calibration across multiple projects. Contributed to adversarial and instruction-following prompt workflows used for safety fine-tuning. • Completed 1,800+ RLHF tasks with a 97% quality acceptance rate. • Designed annotator guidelines for prompt diversity and bias detection across 12 active projects. • Red-teamed frontier LLMs and produced 400+ structured adversarial prompts adopted into safety fine-tuning pipelines. • Mentored 15 junior trainers, reducing team error rates by 34%.

2022 - Present

AI Data Specialist (Contract), Appen

AppenTextClassification

Annotated multimodal training data for machine learning pipelines to support Fortune 500 AI products. Expanded and maintained taxonomies to improve label coverage while preventing category overlap. Applied annotation calibration practices including inter-annotator agreement scoring. • Labeled 50,000+ data points across image, text, audio, and video modalities. • Led a taxonomy expansion project improving label coverage by 22%. • Authored best-practice documentation for IAA scoring and calibration protocols. • Ensured consistent labeling quality across annotators for ML training datasets.

2021 - 2022

Freelance Prompt Engineer & NLP Annotator, Independent

OtherTextPrompt + Response Writing (SFT)

Collaborated with early-stage AI teams to build training datasets for conversational AI, chatbots, and text classification. Produced instruction-following prompt-response pairs and supported dataset creation for model instruction adherence improvements via A/B testing. Focused on dataset quality and practical usability of labeled conversational content. • Built training datasets for conversational AI and text classification systems. • Specialized in instruction-following datasets and diverse prompt-response pair creation. • Improved instruction adherence in A/B testing through curated training examples. • Supported early-stage product dataset needs for NLP model development.

2019 - 2021

Education

Work History