Mincheng K.

Key Skills

Software

Don't disclose

Top Subject Matter

Scientific logic tutoring

LLM reasoning verification via structured inquiry frameworks

LLM prompt engineering

Top Data Types

Text

Top Task Types

Question Answering

RLHF

Freelancer Overview

Senior Science Educator & Multi-Disciplinary Logic Specialist (AI-style logic tutoring and reasoning verification). Brings 13+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose, Internal, and Proprietary Tooling. Education includes Bachelor of Science, National Taichung University of Education (NTCU) and TOEIC Gold Certificate, N/A. AI-training focus includes data types such as Text and labeling workflows including Question Answering and RLHF.

Labeling Experience

Proprietary Financial and Systematic Trader - Self-Employed

TextRLHF

Operates as a proprietary financial and systematic trader, developing and executing algorithmic strategies through rigorous analysis and continuous refinement. Manages cross-border trading and account operations while applying risk management controls under volatile market conditions. Automates parts of operational scheduling and account monitoring using prompt-driven workflow scripts to reduce administrative overhead. • Engineered trading strategies with backtesting, statistical analysis, and ongoing optimization • Managed cross-border financial accounts and funding velocity across international platforms • Applied high-pressure risk management principles to protect capital efficiency • Built automated workflow scripts for scheduling and monitoring to cut recurring administrative time

2025 - Present

Proprietary Financial & Systematic Trader (Self-Employed) — AI collaboration and RLHF dataset methodology

TextRLHF

Developed prompt-driven workflows and methodology to generate rigorous, structured datasets intended for RLHF-style training and evaluation. Leveraged chain-of-thought (CoT) logic engineering with iterative scientific verification cycles to turn benchmarks into zero-error step-by-step reasoning. Audited and rewrote model outputs for culturally localized, scientifically precise, structured qualitative data suitable for downstream AI training. • Built iterative verification loops for prompt structures and response quality • Conducted red-team style adversarial testing to identify and correct errors • Localized and structured multi-sensory flavor profiling content (Traditional Chinese) • Automated prompt-driven operations and monitoring to streamline dataset curation

2025 - Present

Senior Science Educator & Multi-Disciplinary Logic Specialist (AI-style logic tutoring and reasoning verification)

Don't discloseTextQuestion Answering

Designed and delivered AI-tutoring style instruction that audits and refines structured reasoning chains using scientific inquiry cycles. Produced and enforced question/answer frameworks (Problem Identification ➝ Hypothesis ➝ Experimental Design ➝ Empirical Verification) to improve learner outputs toward correct, verifiable reasoning. Conducted adversarial Socratic questioning and verification loops to reduce reasoning errors in student-generated solutions. • Used scientific inference cycles to structure prompts and reasoning verification • Applied adversarial questioning to elicit and correct open-ended hypotheses • Audited logical thought chains against empirical verification criteria • Built cross-disciplinary knowledge graph connections to support response correctness

2014 - Present

Education

Work History