For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Olga Solomon

Olga Solomon

Data analyst & ML QA Group Manager

Israel flagAlfey Menashe, Israel
$22.00/hrExpertInternal Proprietary Tooling

Key Skills

Software

Internal/Proprietary Tooling

Top Subject Matter

AI/ML Data Quality and Annotation

Top Data Types

DocumentDocument
ImageImage
TextText
VideoVideo

Top Task Types

Classification
Data Collection
Entity Ner Classification
Evaluation Rating
Mapping

Freelancer Overview

Data analyst & ML QA Group Manager. Core strengths include Internal and Proprietary Tooling. Education includes Bachelor's Degree in Education, Omsk State University (2016). AI-training focus includes data types such as Document and labeling workflows including Evaluation and Rating.

ExpertHebrewEnglishRussian

Labeling Experience

Semi-Structured Extraction & Table Classification Evaluation (Financial Data)

Internal Proprietary ToolingDocumentEntity Ner Classification
Led evaluation and quality validation of semi-structured extraction models focused on extracting financial information from tables and classifying table types within financial documents. Validated model outputs against expected schemas, ensuring correctness of field extraction, table classification, and value normalization. Identified and categorized extraction errors, including missing fields, incorrect mappings, misclassified tables, structural misalignment, and formatting-related failures. Performed systematic error analysis to detect recurring patterns and edge cases, assessed their impact on downstream usage, and produced structured evaluation reports with clear prioritization of issues. Provided actionable recommendations to improve model logic, training data, and extraction rules, contributing to measurable improvements in extraction accuracy and coverage. Worked closely with Product, Engineering, and SMEs to feed evaluation insights into model iterations and retraining cycle

Led evaluation and quality validation of semi-structured extraction models focused on extracting financial information from tables and classifying table types within financial documents. Validated model outputs against expected schemas, ensuring correctness of field extraction, table classification, and value normalization. Identified and categorized extraction errors, including missing fields, incorrect mappings, misclassified tables, structural misalignment, and formatting-related failures. Performed systematic error analysis to detect recurring patterns and edge cases, assessed their impact on downstream usage, and produced structured evaluation reports with clear prioritization of issues. Provided actionable recommendations to improve model logic, training data, and extraction rules, contributing to measurable improvements in extraction accuracy and coverage. Worked closely with Product, Engineering, and SMEs to feed evaluation insights into model iterations and retraining cycle

2020 - 2021

NER & Text Classification Model Evaluation and Error Analysis

Internal Proprietary ToolingTextEntity Ner Classification
Performed in-depth evaluation of NER and text classification model outputs against gold-labeled datasets. Validated model predictions, systematically identified errors, and classified mistakes into structured categories and issue types (e.g., boundary errors, wrong entity type, missed entities, misclassification across taxonomy levels). Conducted quantitative and qualitative analysis to assess error patterns and their business impact, including calculation of POR (Precision over Recall) and other relevant quality metrics. Produced detailed evaluation reports with breakdowns of recurring issue patterns, root-cause analysis, and clear recommendations for model improvement, data refinement, and guideline updates. Worked closely with Product, Engineering, and SMEs to feed findings back into model retraining cycles and annotation guideline improvements.

Performed in-depth evaluation of NER and text classification model outputs against gold-labeled datasets. Validated model predictions, systematically identified errors, and classified mistakes into structured categories and issue types (e.g., boundary errors, wrong entity type, missed entities, misclassification across taxonomy levels). Conducted quantitative and qualitative analysis to assess error patterns and their business impact, including calculation of POR (Precision over Recall) and other relevant quality metrics. Produced detailed evaluation reports with breakdowns of recurring issue patterns, root-cause analysis, and clear recommendations for model improvement, data refinement, and guideline updates. Worked closely with Product, Engineering, and SMEs to feed findings back into model retraining cycles and annotation guideline improvements.

2017 - 2018

Named Entity Recognition (NER) Labeling & Entity Normalization for News Data

Internal Proprietary ToolingDocumentEntity Ner Classification
Labeled and validated training and test datasets for Named Entity Recognition (NER) models using news documents. Manually annotated entities such as companies, persons, geographic locations, and monetary amounts within unstructured text. Performed entity normalization, ensuring that multiple mentions and aliases referring to the same real-world entity were consolidated into a single normalized entity representation. Followed strict annotation guidelines and schemas to ensure consistency across datasets, contributing to high-quality model training and evaluation.

Labeled and validated training and test datasets for Named Entity Recognition (NER) models using news documents. Manually annotated entities such as companies, persons, geographic locations, and monetary amounts within unstructured text. Performed entity normalization, ensuring that multiple mentions and aliases referring to the same real-world entity were consolidated into a single normalized entity representation. Followed strict annotation guidelines and schemas to ensure consistency across datasets, contributing to high-quality model training and evaluation.

2016 - 2017

Classification Taxonomy & Labeling Guidelines Design

Internal Proprietary ToolingTextClassification
Financial Intelligence, Business News, NLP / AI Training

Financial Intelligence, Business News, NLP / AI Training

2016 - 2016

Data analyst & ML QA Group Manager

Document
Led and managed QA and annotation operations for ML training data in various projects. Oversaw document-based data labeling and quality verification by data analysts and editors. Ensured high-quality datasets for downstream AI model training. • Implemented rigorous QA processes for annotation quality • Coordinated annotation staff and resolved data ambiguities • Shaped annotation guidelines with data scientists • Ensured project timelines and data delivery standards.

Led and managed QA and annotation operations for ML training data in various projects. Oversaw document-based data labeling and quality verification by data analysts and editors. Ensured high-quality datasets for downstream AI model training. • Implemented rigorous QA processes for annotation quality • Coordinated annotation staff and resolved data ambiguities • Shaped annotation guidelines with data scientists • Ensured project timelines and data delivery standards.

2011 - 2016

Education

O

Omsk State University

Bachelor's Degree in Education, Education, English and German

Bachelor's Degree in Education
2011 - 2016
O

Omsk State Pedagogical University

BA, Education

BA
1992 - 1997

Work History

L

LSEG

Solution Manager

Petach Tikva
2020 - 2025
R

Refinitiv

Customer Success Management Team Lead

Petach Tikva
2018 - 2020