Qing Z.

Key Skills

Software

No software listed

Top Subject Matter

LLM evaluation

Rlhf Domain Expertise

and response quality labeling (bilingual Chinese-English)

Top Data Types

Text

Audio

Image

Top Task Types

RLHF

Audio Recording

Freelancer Overview

Contributing Writer (Outlier) — LLM & AI training / evaluation. Brings 16+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. Education includes Diploma of Education, Western Community College (2023) and Diploma, Capilano University (2018). AI-training focus includes data types such as Text and Audio and labeling workflows including RLHF and Audio Recording.

Labeling Experience

Contributing Writer (Outlier) — TTS & audio-related AI training work

AudioAudio Recording

Contributed to TTS and Mandarin voice-related projects involving speech data preparation and quality checks for AI speech systems. Verified pronunciation accuracy, supported speech data collection, and performed localization review to ensure linguistic correctness and naturalness. Helped optimize conversational naturalness for speech-based models through iterative corrections and validation. • Pronunciation accuracy validation for Mandarin voice data • Speech data collection and localization review • Naturalness optimization for AI speech systems • Prompt/speech-model response refinement for improved conversational quality

2024 - Present

Contributing Writer (Outlier) — LLM & AI training / evaluation

TextRLHF

Worked on bilingual Chinese-English LLM and AI training evaluation tasks, primarily involving RLHF-related dataset labeling and reviewer alignment. Developed rubrics and evaluation criteria, then assessed model outputs for factuality, reasoning, safety, tone, and instruction adherence. Produced high-quality “golden responses” for benchmark calibration and improving consistency across review processes. • RLHF, response evaluation, and quality assurance for large language models • Rubric/evaluation-criteria development and output scoring • Golden-response creation for benchmark calibration and reviewer alignment • Multilingual scripts, linguistic correction, and response rewriting to improve conversational quality

2024 - Present

Education

Work History