Stanislav S. - Data Annotator

Lviv, Ukraine

Experience

Dec 2024 - May 2025

6 months

Lviv, Ukraine

Annotate and evaluate mathematics content in Ukrainian, Russian, and English languages
Solve and analyze unsolved problems using basic high school-level mathematics
Work with various math topics including Calculus, Algebra and Functions, Data, Statistics and Probability, Geometry and Structured Images
Analyze tables, quantitative diagrams, infographics, and image-based documents to create high-quality and engaging math questions
Maintain accuracy, clarity, and detail in all annotations and QA development
Collaborate with an international team to ensure content meets educational standards
Utilize RLHF on Turing platform
Apply time management, critical thinking, and attention to detail

Oct 2024 - Dec 2024

3 months

Lviv, Ukraine

Analyze and label linguistic data such as text, audio, or speech according to project guidelines
Identify and correct grammatical, syntactic, and semantic errors
Evaluate and annotate language content to train and improve AI and NLP models
Follow detailed instructions and maintain high annotation accuracy and consistency
Collaborate with team members and provide feedback to enhance annotation quality
Use Google Sheets, MS Office, and Excel

Jul 2019 - Aug 2023

4 years 2 months

Poltava, Ukraine

Assist in rigging-up/rigging-down, equipment inspections, and preventive maintenance
Perform pre-job calculations (volumes, pressures, capacities) and prepare technical documentation
Execute cementing and plug jobs including pressure testing up to 15,000 PSI
Ensure compliance with safety standards and accurate reporting

Nov 2018 - May 2019

7 months

Poltava, Ukraine

Assist in rigging-up and rigging-down of rig equipment
Perform general repairs and maintenance on major and minor rig components
Assist with mud-mixing duties
Perform catching samples, use tongs, slips, and chain to connect and disconnect drilling pipe, tubulars, and drilling bits or tools
Check hoist lines and ropes for blocks

Evaluated quality of potentially unethical or unsafe LLM model prompts and responses translated into Russian
Assessed translation quality, response quality, and model’s ability to deflect unsafe material while remaining relevant to prompts

Annotated images containing English and Russian words with bounding boxes at word, line, and paragraph levels using SRT HALO tool
Transcribed printed and handwritten text with high accuracy preserving case, accents, and special characters
Identified languages per word and flagged illegible or unknown scripts
Differentiated between overlaid and scene text to improve OCR training data quality
Ensured consistency by correcting pre-annotations and following rejection criteria

Annotated and validated over 10,000 Ukrainian text samples for dialect identification in GenAI datasets
Corrected system-predicted language labels including Romanized variants and unsupported dialects
Applied quality standards to reject non-linguistic content such as gibberish, emojis, hashtags, and proper nouns
Ensured accuracy and consistency of annotations across multiple dialects and scripts