Irena Georgieva

Key Skills

Software

Google Cloud Vertex AI

OneForma

Toloka

Top Subject Matter

No subject matter listed

Top Data Types

Text

Video

Top Task Types

Bounding Box

Freelancer Overview

I have hands-on experience in AI training and data labeling across diverse domains, including education, e-commerce, and linguistics. My work has involved annotating text for sentiment analysis, categorizing product information, and labeling sensitive content with a high degree of accuracy. I am proficient in using tools such as Labelbox, Prodigy, and Scale AI, and I prioritize consistency, quality, and data integrity in every task. My strong attention to detail and ability to quickly adapt to domain-specific guidelines make me a dependable contributor to the development of high-performing machine learning models.

Labeling Experience

RWS Project

TolokaImageBounding Box

I contributed to a large-scale project focused on political discourse analysis, where the goal was to train models to detect sentiment, bias, stance, and topic relevance in political texts. My responsibilities included labeling social media posts, news excerpts, and political speeches for emotional tone, ideological slant, and factual content. The project involved tens of thousands of samples and required strict adherence to quality control measures, including inter-annotator agreement thresholds and periodic calibration tasks. I maintained high accuracy and consistency while working with nuanced and often subjective political content, helping ensure the reliability of downstream NLP applications.

I contributed to a large-scale project focused on political discourse analysis, where the goal was to train models to detect sentiment, bias, stance, and topic relevance in political texts. My responsibilities included labeling social media posts, news excerpts, and political speeches for emotional tone, ideological slant, and factual content. The project involved tens of thousands of samples and required strict adherence to quality control measures, including inter-annotator agreement thresholds and periodic calibration tasks. I maintained high accuracy and consistency while working with nuanced and often subjective political content, helping ensure the reliability of downstream NLP applications.

2024 - 2024

Education

Work History