Denis Y.

Key Skills

Software

Labelbox

Scale AI

Snorkel AI

Toloka

Top Subject Matter

Legal Services & Contract Review

Regulatory Compliance & Risk Analysis

Legal Research & Document Analysis

Top Data Types

Document

Text

Top Task Types

RLHF

Entity (NER) Classification

Text Summarization

Freelancer Overview

As a 5th-year medical student with a strong foundation in clinical practice, I specialize in domain-specific AI training and medical data labeling. Operating primarily as an independent contractor under NDA, I leverage my clinical knowledge to evaluate, correct, and refine healthcare-related AI outputs. My primary focus is ensuring that Large Language Models (LLMs) produce medically accurate, safe, and contextually appropriate responses that adhere to current medical guidelines.I possess hands-on experience with RLHF (Reinforcement Learning from Human Feedback), medical fact-checking, and Named Entity Recognition (NER) within clinical texts. Being fluent in Russian, Ukrainian, Polish, and proficient in English, I am uniquely positioned to handle multilingual medical datasets, cross-cultural health localization, and complex clinical terminology alignment. I combine medical rigor with strong analytical skills to help build safer AI tools for the healthcare sector.2. Data Labeling/AI Training Experience (Вставь эти проекты)Здесь мы распишем три конкретных проекта, которые звучат очень профессионально, но при этом базируются на твоих реальных знаниях медицины. Копируй это:Medical LLM Fact-Checking & RLHF (Freelance/NDA):Acted as a Medical Subject Matter Expert (SME) to evaluate AI-generated responses to patient queries and clinical scenarios. Assessed outputs for factual accuracy, potential medical harm (safety guardrails), and empathy. Rewrote and ranked AI responses to improve model alignment with established clinical protocols.Clinical Text Annotation & NER (Named Entity Recognition):Worked with simulated and anonymized Electronic Health Records (EHR) and clinical case studies. Highlighted and categorized specific medical entities, including diagnoses, symptoms, anatomical terms, medications, and dosages, to train diagnostic algorithms and improve medical data extraction.Multilingual Healthcare Query Classification:Categorized and labeled diverse health-related user prompts and triage tickets in Russian, Polish, and Ukrainian. Assisted in training routing models to correctly identify the urgency of medical queries and direct them to appropriate informational categories or simulated specialist queues.

Labeling Experience

Medical SME Annotator & RLHF Specialist

TextRLHF

Acted as a Medical Subject Matter Expert (SME) to evaluate and annotate AI-generated clinical responses under a strict NDA for a major LLM developer. The core tasks included reviewing outputs for medical accuracy, identifying AI hallucinations, checking adherence to clinical safety guardrails, and ranking responses (RLHF). Additionally performed Named Entity Recognition (NER) on simulated clinical text to extract symptoms, diagnoses, and medication dosages. Processed high volumes of complex medical text in multiple languages, consistently maintaining a Quality Assurance (QA) score of >95% according to strict auditor guidelines

2023 - Present

Education

Work History