Diana Ivanova

Analytical multilingual AI evaluation specialist with hands-on experience reviewing AI-generated content, multilingual data annotation, linguistic quality assurance (LQA), transcription verification, and guideline-driven quality assessment. Experienced in evaluating model accuracy, naturalness, grounding, consistency, formatting, and cultural appropriateness. Comfortable comparing AI responses side-by-side (SxS), identifying hallucinations, writing evidence-based evaluation rationales, and delivering high-quality work independently in remote environments.

AI & Language Specialist - Freelance - Remote

(2025-01)

• Evaluated AI-generated conversations for accuracy, clarity, consistency, grammar, and natural language quality.
• Performed multilingual AI data annotation supporting model training and quality improvement.
• Compared AI responses side-by-side (SxS), identified grounding issues and hallucinations, and produced structured, evidence-based rationales.
• Reviewed transcription and annotation outputs while following detailed project guidelines and quality standards.
• Applied linguistic judgement to improve semantic accuracy, readability, and cultural appropriateness across multiple languages.
• Worked independently on remote AI projects while consistently meeting quality expectations and deadlines.

GESE Grade 5 - Trinity College London (2026-04)