Tingting S.

Key Skills

Software

Other

OpenCV AI Kit (OAK)

Labelbox

AWS SageMaker

Top Subject Matter

Document OCR and detection for financial and identity documents

Financial document extraction and multi-line amount accuracy improvement

Training data collection

Top Data Types

Document

Text

Image

Top Task Types

Text Summarization

Data Collection

Freelancer Overview

Independent data annotation contributor for document OCR/detection dataset creation. Brings 18+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Other, N, and A. Education includes Master of Education, University of Massachusetts Amherst (2009) and Bachelor of Arts, North University of China (2005). AI-training focus includes data types such as Document and Text and labeling workflows including Evaluation, Rating, and Text Summarization.

Labeling Experience

Annotation guidance and labeling platform quality feedback role

Text

Collaborated across data, product, and engineering teams to author user guides for data annotation and labeling content delivered to customers. Conducted quality testing and provided feedback on annotation and training engines and platforms used for AI and NLP workflows. Ensured labeling guidance aligned with customer-facing requirements and system performance. • Wrote and maintained user guides for annotation workflows • Provided quality testing and performance feedback on labeling engines • Coordinated with multi-disciplinary teams (data/product/engineering) • Supported customer delivery of labeling and training guidance

Not specified

Training data preparation and synthetic data generation contributor

OpenCV AI Kit (OAK)DocumentData Collection

Used data crawling to collect varied training data and transformed unstructured content into key-value pairs to support model training. Generated synthetic training data using computer vision models and tooling, including tax forms and payroll-related materials. Assisted in converting and structuring data so it could be consumed by AI training pipelines. • Crawled and collected heterogeneous training datasets • Normalized unstructured data into key-value representations • Generated synthetic data for forms such as taxes and payrolls • Prepared structured inputs to improve training effectiveness

Not specified

Cross-functional work supporting labeled-data services for fund contracts

DocumentText Summarization

Converted business requirements into project-oriented data preparation and customized labeling services for fund contract content. Improved accuracy for multi-line amount expressions used in labeling and downstream extraction tasks. Coordinated with engineering teams to support ETL pipelines for preparing data used by training and annotation engines. • Supported customized extraction functions for fund contract documents • Improved multi-line amount accuracy requirements • Collaborated with business and engineering teams on data preparation • Contributed to labeling service improvements for customer needs

Not specified

Independent data annotation contributor for document OCR/detection dataset creation

OtherDocument

Labeled more than 20,000 documents with high accuracy, including ID cards, bank cards, bank statements, utility bills, and fund contract documents. Collaborated with an AI team to train high-precision machine learning models focused on detection and OCR performance. Quality testing and feedback were provided on annotation and training engines and platforms (including real-time AI and NLP engines). • Annotated document fields/records for OCR and detection tasks • Trained/validated ML models using labeled datasets • Achieved detection and OCR accuracy rates up to 99% • Performed quality assurance and iteration feedback loops for training pipelines

Not specified

Education

Work History