王帅

Key Skills

Software

Other

Top Subject Matter

Chinese AI training and Simplified Chinese LLM evaluation

educational content quality review

Mandarin speech transcription QA

Top Data Types

Text

Audio

Top Task Types

Transcription

Segmentation

RLHF

Entity (NER) Classification

Classification

Freelancer Overview

Freelance Chinese AI Annotation, Content Review, and Educational Rewriting (LLM/RLHF-style response evaluation + Chinese. Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Bachelor of Arts, Harbin Cambridge University (2023) and Mandarin Proficiency Certificate, Level 2-A, N/A. AI-training focus includes data types such as Text and Audio and labeling workflows including Evaluation, Rating, and Transcription.

Labeling Experience

Freelance Chinese Content Reviewer and Educational Rewriter - Freelance

TextSegmentationRLHF

Freelanced remotely to evaluate Chinese AI outputs and ensure they meet task requirements for relevance, completeness, and factual accuracy. Produced and quality-checked Simplified Chinese text datasets and educational materials while following detailed guidelines and edge cases. Applied native language judgment and teaching-oriented clarity to judge tone, naturalness, and usefulness for end users. • Evaluated instruction-following, intent alignment, and helpfulness • Labeled and reviewed Simplified Chinese text for quality and error identification • Rewrote and polished educational questions and answer explanations • Compared outputs across AI tools to improve prompts and assess usability

2024 - Present

Freelance Mandarin speech transcription and transcript QA (as part of Chinese AI training content review)

OtherAudioTranscription

Reviewed Mandarin speech/transcript-related quality aspects such as punctuation, segmentation, and naturalness cues when applicable to tasks. Applied native Simplified Chinese and Mandarin proficiency to judge whether audio-derived text matched standard usage and expected formatting. Ensured outputs were suitable for educational and evaluation purposes. • Checked transcript punctuation and segmentation consistency with Mandarin conventions. • Judged pronunciation/naturalness indicators when reviewing Mandarin text tied to speech tasks. • Verified clarity and logical coherence of transcript-derived content. • Identified guideline edge cases affecting readability and learner comprehension.

2024 - Present

Freelance Chinese AI Annotation, Content Review, and Educational Rewriting (LLM/RLHF-style response evaluation + Chinese text content QA)

OtherText

Evaluated Chinese AI responses for relevance, completeness, factual consistency, tone, and user-intent fit according to task guidelines. Performed quality scoring and error identification to judge instruction-following and hallucination risk in model outputs. Conducted guideline-based Chinese language judgments to determine whether responses were usable for real needs. • Rated response quality and helpfulness using criteria such as clarity, naturalness, and usefulness. • Checked for factual consistency and potential hallucination/safety issues. • Assessed instruction-following, relevance, and intent alignment against the user request. • Compared outputs across multiple LLM tools to identify weak reasoning and inform prompt refinement.

2024 - Present

Education

Work History