Yao Zhong Z.

Key Skills

Software

Other

Top Subject Matter

Cross-border e-commerce operations and AI logic audit/training workflows

Cross-border e-commerce operations and knowledge-to-training instruction transformation

Top Data Types

Text

Document

Top Task Types

RLHF

Prompt + Response Writing (SFT)

Text Summarization

Classification

Question Answering

Freelancer Overview

Plato-2 Cognitive System Construction based on Obsidian (System Architect / Data Annotation Supervisor). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Google Stitch 2.0. Education includes Associate Degree, N/A. AI-training focus includes data types such as Text and Document and labeling workflows including RLHF and Prompt + Response Writing (SFT).

Labeling Experience

Plato-2 Cognitive System Construction based on Obsidian (System Architect / Data Annotation Supervisor)

TextRLHF

Designed an Obsidian-based AI cognitive system and produced high-quality annotation assets for RLHF training through conflict mediation and synthesis. Manually annotated and corrected AI-generated cognitive conflicts within audit closed loops to enforce single source of truth standards. Built structured JSON Canvas mediation workflows that generate integrative business-logic samples for model learning.• Implemented manual annotation/correction of cognitive conflicts during 81 audit closed loops• Established decision/log discrimination standards to maintain SSoT• Designed visual conflict mediation using JSON Canvas and collaboration between scan and relationship agents• Developed system guardrails (e.g., GEMINI.md) to reduce hallucination and logic deviations

2026 - Present

Cross-border E-commerce Website Operations & Automated Delivery (E-commerce Operations Manager / Full-stack Delivery Expert)

DocumentPrompt + Response Writing (SFT)

Translated e-commerce operational processes into executable manuals and structured knowledge that can be converted into SFT training data and annotation rules. Standardized advertising placement logic and supply-chain management procedures to produce reusable instructional content for AI-guided scenarios. Used cross-border e-commerce context and Chinese language logic to improve semantic precision in training-ready guidance.• Standardized complex advertising placement logic into instructional documentation• Converted operational scripts (including customer-service standards) into structured notes for AI training use• Integrated Google Stitch 2.0 and AntiGravity IDE to deliver automated scenario workflows• Improved organic traffic and conversion by injecting language-logic structure into SEO guidance

2025 - Present

Education

Work History