现实 吕.

Key Skills

Software

Other

Prodigy

Top Subject Matter

AI training data preparation (LLM interactions)

Top Data Types

Text

Image

Video

Top Task Types

Prompt + Response Writing (SFT)

Freelancer Overview

Familiar with interaction logic and usage boundaries of mainstream LLMs (GPT, Claude, Wenxin Yiyan); understand prompt design and output evaluation from a training data quality perspective. Strong Chinese writing and logical decomposition skills; able to efficiently complete annotation tasks including instruction understanding, answer rewriting, and relevance judgment. Highly attentive to data quality and consistency; capable of producing high-quality training samples that strictly follow annotation guidelines. Foundational understanding of AI alignment and RLHF; committed to continuous learning through real-world AI trainer work

Labeling Experience

University practical projects (AI training oriented, no formal job experience)

OtherTextPrompt + Response Writing (SFT)

Worked on AI training-oriented practical projects during university, focusing on creating and interpreting high-quality training content. Applied strong instruction comprehension and logical reasoning to produce training data aligned with given requirements. Iterated on outputs to improve clarity and usefulness for AI interaction logic. • Interpreted requirements and translated them into training-friendly prompts • Produced and refined prompt/response pairs to support SFT-style training • Performed basic data preparation through cleaning and categorization • Communicated outputs clearly in written Chinese to match instruction intent

2024 - Present

Education

Work History