刘洋

Key Skills

Software

AWS SageMaker

Top Subject Matter

Chinese public sector

smart community

government digital transformation

Top Data Types

Text

Document

Top Task Types

Entity (NER) Classification

Prompt + Response Writing (SFT)

Text Generation

Text Summarization

Evaluation/Rating

Freelancer Overview

I have hands-on project experience in public-security data platform support, especially around AI training data readiness and data quality preparation. In a smart community / public-security data platform project, I participated in metadata profiling, data-source discovery, data collection coordination, data cleansing review, data standardization, and governance validation. My work involved clarifying field meanings with business stakeholders, reviewing sample data quality, identifying missing or inconsistent records, mapping business workflows into structured data requirements, and supporting acceptance checks for downstream analytics and AI-enabled application scenarios. Although I was not responsible for model algorithm development, my experience is highly relevant to AI training work because high-quality AI outputs depend on accurate, well-structured, well-governed domain data. I am familiar with reviewing Chinese public-sector data, understanding business semantics, checking data consistency, and validating whether datasets are usable for annotation, evaluation, and AI workflow support. Education includes Associate Degree, Guizhou Vocational Technology College (2007). AI-training focus includes data types such as Text and labeling workflows including Evaluation and Rating.

Labeling Experience

AI Training & LLM Evaluation Specialist (Public-Sector AI Business Orchestration)

Text

Evaluated and reviewed LLM and AI-generated outputs in Chinese for workflow correctness, risk-control logic, and user intent alignment. Applied rubric-based QA to judge the quality of AI-assisted business orchestration outputs in public-sector smart community implementations. Ensured structured and accurate responses from LLMs by validating against operational and acceptance-quality standards. • Extensive experience reviewing Chinese domain-specific LLM outputs. • Focused on workflow logic, intent, and business-acceptance criteria. • Participated in prompt iteration and guideline-based evaluation. • Conducted quality review of AI model responses for smart community scenarios.

2020 - 2024

Education

Work History