Corina - Gabriela Dragoi

I’m an AI evaluation and data annotation professional with a computer science and cybersecurity background. I assess AI-generated text, image, and audio content against detailed rubrics, identifying hallucinations and instruction-following issues while writing clear quality rationales.

I bring strong evaluator judgment and a neutral, policy-grounded approach to ambiguous cases. I focus on factual consistency, localization quality, safety review, and formatting/edge-case accuracy—always supported by structured feedback categorization.

Since 2025, I’ve worked as an AI Training Specialist for remote projects at Centific AI, comparing model outputs and judging response quality through rubric-based workflows. I adapt quickly to changing guidelines, annotation tools, and QA quality targets, and I consistently flag missed instructions, unsafe content, and localization problems with well-grounded notes.

In parallel, I’ve contributed as an AI Evaluator for Mindy Support, supporting large-scale text/image/audio evaluation efforts. My technical foundation includes IT support and access administration work, and I leverage that process discipline alongside multilingual strengths in Norwegian and English localization review.