Liqing T.

Key Skills

Software

Labelbox

Prodigy

Top Subject Matter

Advanced Software Engineering & AI Code Optimization

Cybersecurity & Vulnerability Assessment (OWASP)

Complex Logic Verification & Algorithmic Reasoning

Top Data Types

Computer Code Programming

Text

Document

Top Task Types

Prompt + Response Writing (SFT)

RLHF

Question Answering

Text Generation

Text Summarization

Red Teaming

Function Calling

Freelancer Overview

Senior Full-Stack Technical Expert (L7) with over 10 years of experience in high-concurrency systems and software architecture. Specialized in Code Quality Auditing and AI-Assisted Development. I have a deep understanding of RLHF principles, specifically in evaluating LLM-generated code for logical correctness, security vulnerabilities, and architectural efficiency. Expert in TypeScript, and distributed systems.

Labeling Experience

AI-Driven Development Architect — Terminus Technologies (AI automation integration with AI-auditor evaluation of code outputs)

Computer Code ProgrammingPrompt + Response Writing (SFT)

Evaluated and refined thousands of AI-generated code snippets to ensure production-grade robustness within an AI-auditor workflow integrated into the SDLC. The work emphasized consistency checking, automated error correction, and verification of logic before code was considered reliable. This aligns with training/evaluation of LLM outputs by judging quality and grounding corrections against “production” standards. • Reviewed large volumes of model-generated programming artifacts. • Audited logic consistency and identified issues in JS/TS code paths. • Supported automated correction and verification steps in an end-to-end pipeline. • Reduced iteration cycles by enforcing quality gates on AI outputs.

2023 - Present

Senior Code Auditor & Technical Expert — Alibaba Group (L7) (human evaluation/review to support RLHF-quality coding standards)

Computer Code ProgrammingRLHF

Led high-stakes code reviews for core DevOps and infrastructure components and provided high-precision feedback on logic, security vulnerabilities, and performance. The feedback was explicitly framed as directly usable for High-Value RLHF for coding models to ensure outputs meet enterprise engineering standards. This constitutes human evaluation and rating of AI-generated programming behavior and risk characteristics. • Delivered detailed security and correctness feedback for large-scale services. • Audited for XSS/CSRF and other security-relevant issues. • Assessed performance characteristics and architectural consistency. • Applied deep debugging and verification methods to guide model improvement.

2019 - 2023

Education

Work History