Nhat A.

Key Skills

Software

Other

Data Annotation Tech

Datumbox

Datature

Top Subject Matter

Content creation

digital marketing

data analysis

Top Data Types

Text

Video

Computer Code Programming

Image

Top Task Types

Text Generation

Classification

Prompt + Response Writing (SFT)

Freelancer Overview

AI Response Rating and LLM Output Evaluation. Brings 1+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other, Internal, and Proprietary Tooling. Education includes Bachelor of Science, 长沙理工大学 (2028). AI-training focus includes data types such as Text and Image and labeling workflows including Evaluation, Rating, and Classification.

Labeling Experience

Prompt Writing and AI Response Instruction

OtherTextPrompt + Response Writing (SFT)

I designed, drafted, and iterated prompts to guide AI models toward specific response objectives. Through systematic prompt engineering, I improved output relevance and adherence to task requirements. My work contributed to supervised fine-tuning and prompt effectiveness evaluation. • Created targeted prompts for various LLMs • Conducted iterative prompt testing and refinement • Evaluated output alignment with task criteria • Supported multilingual and multi-domain tasks

2024 - Present

Image Annotation and Multimodal Data Labeling

ImageClassification

I performed image content identification, categorization, and labeling for e-commerce and AI-driven image tasks. The work included following detailed annotation guidelines and optimizing visual data for business use. I also supported multimodal data processing within proprietary platforms. • Labeled e-commerce product images and generated image sets • Applied rule-based procedures for accurate annotations • Participated in quality control and task review cycles • Supported integration with AI-powered generation tools

2024 - Present

AI Response Rating and LLM Output Evaluation

OtherText

In this role, I evaluated AI-generated responses for accuracy, relevance, completeness, and clarity using major language models such as ChatGPT, Claude, Gemini, DeepSeek, and Tongyi Qianwen. I compared outputs from different models to assess strengths and weaknesses in both Chinese and English tasks. Additionally, I wrote prompts and judged response quality for various domains. • Performed multi-criteria evaluation of LLM outputs • Identified issues like irrelevance, factual errors, and poor expression • Designed task-oriented prompts for optimal model output • Used evaluations to inform further model and prompt iteration.

2024 - Present

Education

Work History