暮花流水

Key Skills

Software

Other

Top Subject Matter

AI Agent Simulation

Job Search/Natural Language Processing

Recommendation System

Top Data Types

Text

Top Task Types

Prompt + Response Writing (SFT)

Entity (NER) Classification

Freelancer Overview

Machine learning analyzer. Brings 4+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Master of Science, University of Washington Seattle (2024) and Bachelor of Science, University of Illinois Urbana-Champaign (2024). AI-training focus includes data types such as Text and labeling workflows including Prompt + Response Writing (SFT) and Entity (NER) Classification.

Labeling Experience

Machine learning analyzer

OtherTextPrompt + Response Writing (SFT)

Replicated a GPT-based AI town simulation and performed modifications to NPC personas and prompt engineering to analyze AI action logic. Created chaos characters via prompt engineering and experimented with AI's conditional probability output preferences. Designed and studied experiments to understand AI weaknesses in reasoning through controlled data labeling and prompt-response structures. • Used prompt engineering techniques to create adversarial and test scenarios for GPT-based models. • Modified and analyzed personalities and behavior outputs for diverse AI agents. • Developed experimental logic to uncover output preferences and weaknesses. • Focused on iterative refinement and SFT-style prompt and response collection for AI reasoning analysis.

2023 - 2024

Software Development Engineering

OtherTextEntity (NER) Classification

Trained a named entity recognition (NER) model targeting job search keywords using the BERT framework. Collected and labeled data through Python scripts and prompt engineering with GPT. Achieved high-accuracy entity extraction and deployed the model end-to-end for real-world keyword analysis. • Labeled and classified job search-related entities in text with supervised learning. • Combined script-driven and manual data annotation methods. • Supported prompt-based and model-labeled dataset expansion for training. • Integrated entity classification into a deployment pipeline for production.

2023 - 2023

Education

Work History