Liu Z.

Key Skills

Software

Label Studio

Top Subject Matter

AI LLM Fine-tuning for Divination (Text Data)

AI Intelligent Document Q&A and Function Calling (Text Data)

Top Data Types

Text

Top Task Types

Fine-tuning

Function Calling

Freelancer Overview

LLM Fine-tuning and Data Annotation Lead (Divination Project). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal and Proprietary Tooling. Education includes Master of Science, University of Strathclyde (2022) and Bachelor of Science, University of International Business and Economics (2017). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning and Function Calling.

Labeling Experience

Data Labeling and Evaluation Specialist (AI Knowledge Base)

TextFunction Calling

Developed document parsing and retrieval pipeline with vectorization and context assembly for intelligent Q&A AI. Labeled and curated training data for function calling tasks, creating question-context-response annotations for LLM evaluation. Conducted hands-on evaluation/ratings of LLM-generated answers with human feedback to improve performance. • Parsed and segmented documents into context chunks for embedding and retrieval testing. • Built function calling datasets for downstream internal API/tool invocation by LLMs. • Labeled ground truth responses and evaluated model output across multiple scenarios. • Performed ongoing annotation and model assessment for knowledge base Q&A and LLM function calls.

2025 - Present

LLM Fine-tuning and Data Annotation Lead (Divination Project)

TextFine-tuning

Built and maintained a fine-tuning data pipeline for LLMs in divination and fortune-telling scenarios. Leads all data cleaning, format alignment, QLoRA fine-tuning, and human-reviewed annotation tasks on a self-developed dataset. Responsible for prompt template engineering and SFT data preparation for divination system prompts and responses. • Processed raw textual data from classical and modern sources, aligning it to Alpaca/ShareGPT formats. • Designed and implemented annotation workflows that included human label review and scoring. • Evaluated fine-tuned models utilizing BLEU, ROUGE, and user satisfaction feedback. • Assembled prompt+response pairs for SFT and function calling tasks, supporting dynamic scenario handling.

2023 - 2024

Education

Work History