Sobrex

Key Skills

Software

Toloka

Top Subject Matter

Natural Language Processing

Large Language Models

Top Data Types

Text

Top Task Types

Transcription

Computer Programming Coding

Prompt Response Writing SFT

Data Collection

Bounding Box

Entity Ner Classification

Freelancer Overview

Freelance AI Training Specialist & Data Annotator. Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Toloka, N, and A. Education includes Bachelor of Science, Covenant University (2025). AI-training focus includes data types such as Text and labeling workflows including Evaluation and Rating.

Labeling Experience

Freelance AI Training Specialist & Data Annotator

TolokaText

As an AI Trainer and Data Annotator, I contributed to high-quality datasets for NLP model training and evaluation. My core responsibilities included labeling, validating, and assessing data points, as well as providing qualitative feedback on AI performance. Adherence to complex guidelines and passing advanced qualification tasks were central to my role. • Labeled and validated over 1,000 text data points for NLP. • Delivered feedback to enhance truthfulness and accuracy of model responses. • Consistently passed exams for 2D and 3D annotation tasks, unlocking complex workflows. • Maintained over 98% accuracy rate using platforms like Toloka and Remotasks.

As an AI Trainer and Data Annotator, I contributed to high-quality datasets for NLP model training and evaluation. My core responsibilities included labeling, validating, and assessing data points, as well as providing qualitative feedback on AI performance. Adherence to complex guidelines and passing advanced qualification tasks were central to my role. • Labeled and validated over 1,000 text data points for NLP. • Delivered feedback to enhance truthfulness and accuracy of model responses. • Consistently passed exams for 2D and 3D annotation tasks, unlocking complex workflows. • Maintained over 98% accuracy rate using platforms like Toloka and Remotasks.

2025 - Present

Self-Published LLM Training/Annotation Researcher

Text

I participated in a self-led project focused on the identification and labeling of hallucinations in LLM outputs. This effort centered around establishing a framework for high-accuracy annotation within edge cases, contributing to reliable LLM performance. Through rigorous application of logical analysis, I ensured consistency and minimized annotation bias. • Developed 99% accuracy framework for detecting LLM hallucinations. • Conducted comprehensive case studies on edge case identification. • Applied annotation strategies to improve model reliability. • Published findings in a self-authored technical document.

I participated in a self-led project focused on the identification and labeling of hallucinations in LLM outputs. This effort centered around establishing a framework for high-accuracy annotation within edge cases, contributing to reliable LLM performance. Through rigorous application of logical analysis, I ensured consistency and minimized annotation bias. • Developed 99% accuracy framework for detecting LLM hallucinations. • Conducted comprehensive case studies on edge case identification. • Applied annotation strategies to improve model reliability. • Published findings in a self-authored technical document.

2025 - 2026

Education

Work History