Key Skills

Software

Other

Top Subject Matter

No subject matter listed

Top Data Types

Medical Dicom

Top Task Types

Object Detection

Point Key Point

Polygon

Polyline

Segmentation

Freelancer Overview

With over a decade of experience in quality assurance for highly complex systems like surgical robots and 5G networks, I have pivoted my expertise to the foundational element of AI: high-quality training data. My unique value lies in applying a rigorous, system-level engineering mindset to the entire AI data pipeline. I don't just manage annotation; I architect the data strategy. For a flagship AI-assisted medical imaging system, I single-handedly defined the pixel-level annotation standards and built the end-to-end quality control process from the ground up, managing a 100,000+ image dataset to an accuracy exceeding 98%. My background in testing allows me to perform deep root-cause analysis on model failures, directly linking data gaps to performance issues. This enabled me to drive targeted data collection that increased model recall for critical cases by 15%. Furthermore, my familiarity with medical compliance (IEC 62304) ensures that the data pipelines I build are not only accurate but also audit-ready for regulatory submissions like NMPA, making me a bridge between raw data and a deployable, trustworthy medical AI product.

Labeling Experience

Education

Work History