Sz L.

Key Skills

Software

Label Studio

CVAT

Scale AI

Snorkel AI

Top Subject Matter

Software Engineering & Code Evaluation

Large Language Models

AI Reasoning

Top Data Types

Text

Video

Computer Code Programming

Top Task Types

Fine-tuning

RLHF

Segmentation

Transcription

Object Detection

Text Generation

Computer Programming/Coding

Freelancer Overview

As a data science graduate and full-stack developer, I view data annotation not merely as a routine task, but as a foundational pillar of artificial intelligence. I possess specialized expertise and experience in Supervised Fine-Tuning (SFT), Reinforcement Learning from Human Feedback (RLHF), and "hallucination" auditing. Leveraging my deep technical background in Python programming and web architecture, I am capable of conducting rigorous and meticulous evaluations of complex code outputs and logical reasoning paths. I am dedicated to providing scrupulous and logically sound evaluation services for Large Language Model (LLM) alignment, cross-lingual translation, and multimodal tasks.

Labeling Experience

AI Agent Trainer / Machine Learning Engineer

TextFine-tuning

At Baidu, I contributed to the fine-tuning of proprietary Large Language Models (LLMs) for targeted business applications. My duties included systematic dataset construction, prompt engineering, and rigorous evaluation of model outputs. I played a key role in addressing model hallucination through careful curation and benchmarking of high-quality evaluation data. • Supervised fine-tuning and RLHF methodologies applied to LLMs • Generated and reviewed training prompts to optimize agent reasoning • Assembled and rated domain-specific evaluation datasets • Participated in model corrections and iterative process improvement

2023 - 2025

Education

Work History