Trang Chu Thị Cẩm

Key Skills

Software

Data Annotation Tech

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Label Types

Translation Localization

Evaluation Rating

Prompt Response Writing SFT

Audio Recording

Transcription

Freelancer Overview

I have hands-on experience as a bilingual AI trainer and data annotator, where I craft, translate, and evaluate prompts for various AI training projects. My work includes assessing model outputs for coherency, fluency, factual accuracy, and safety, as well as determining which model delivers better results. With a background in creative writing, content creation, and translation, I am skilled at producing and refining high-quality training data for NLP and content generation tasks. My expertise in leveraging AI for content strategy and my attention to detail ensure that I consistently deliver accurate and reliable data labeling and annotation for diverse domains.

Labeling Experience

AI Trainer

Scale AITextTranslation LocalizationEvaluation Rating

- Project types: Voice data labeling, mathematical problem solving, conversational prompt evaluation, image annotation - Scope: Large-scale bilingual (EN–VI) language model training projects - Key tasks: +Prompt creation (factual & fictional) across multiple difficulty levels +Model response evaluation based on coherence, fluency, factual accuracy, tone, and relevance +Safety assessment and categorization of guideline-violating content +Prompt translation and localization (EN–VI) - Project size: Thousands of prompts and comparative model evaluations. - Quality: compliance with annotation guidelines, consistency validation, multi-criteria evaluation. The average quality rating of 4/5 across projects.

- Project types: Voice data labeling, mathematical problem solving, conversational prompt evaluation, image annotation - Scope: Large-scale bilingual (EN–VI) language model training projects - Key tasks: +Prompt creation (factual & fictional) across multiple difficulty levels +Model response evaluation based on coherence, fluency, factual accuracy, tone, and relevance +Safety assessment and categorization of guideline-violating content +Prompt translation and localization (EN–VI) - Project size: Thousands of prompts and comparative model evaluations. - Quality: compliance with annotation guidelines, consistency validation, multi-criteria evaluation. The average quality rating of 4/5 across projects.

2023 - 2025

Education

Work History