Haiming Z.

Key Skills

Software

Other

Top Subject Matter

AI Language Model Evaluation

Text Data Annotation for AI

Prompt Engineering & LLM Optimization

Top Data Types

Text

Top Task Types

Classification

Prompt + Response Writing (SFT)

Freelancer Overview

LLM Model Evaluation & Dialogue Annotation. Core strengths include Internal and Proprietary Tooling. AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and Classification.

Labeling Experience

Prompt Writing & Optimization for AI Training

TextPrompt + Response Writing (SFT)

Developed and optimized prompts for LLMs through iterative debugging and analysis to support AI training and model fine-tuning. Customized prompt templates tailored for multiple business scenarios, improving LLM generation quality and output consistency. Provided high-level support for prompt engineering, suitable for advanced annotation and model customization tasks. • Designed reusable prompt templates for diverse use cases. • Improved model output through analytical iteration and evaluation. • Assisted in fine-tuning AI models for client-specific requirements. • Delivered documentation and prompt optimization reports routinely.

2023 - Present

AI Text Data Annotation

TextClassification

Skilled in text classification, sentiment analysis, intent recognition, entity extraction, and multi-turn conversation labeling. Maintained high consistency and accuracy across different annotation projects with proven adaptability. Demonstrated stable working efficiency while quickly adapting to varying project requirements. • Performed sentiment labeling and text categorization across diverse datasets. • Supported complex entity extraction and intent recognition tasks. • Adapted to new guidelines and specifications for each client need. • Ensured consistent, high-quality outputs for project stakeholders.

2023 - Present

LLM Model Evaluation & Dialogue Annotation

Text

Engaged in LLM multi-turn dialogue scoring, logical rationality assessment, factual verification, and discrimination of bias or inappropriate content. Consistently achieved over a 98% personal accuracy rate and 100% on-time delivery rate while following strict annotation standards. Demonstrated rapid mastery of project rules and high-quality data output with minimal rework. • Conducted thorough dialogue annotation and evaluation sessions weekly. • Focused on risk content identification, fact checking, and logical rating. • Maintained stable delivery efficiency and high consistency in output. • Supported LLM prompt assessment and AI model training workflows.

2023 - Present

Education

Work History