I work across AI evaluation, data annotation, preference-based response assessment, prompt engineering practice and Turkish language quality review. My focus is careful guideline interpretation, linguistic accuracy and consistent human evaluation of AI-generated content.
My work combines language expertise, structured evaluation and human judgment. I am especially interested in assessing AI response quality through clear criteria and consistent guideline-based review.
Reviewing and comparing AI-generated responses for relevance, factuality, instruction following, completeness and overall quality.
Evaluating Turkish for naturalness, grammar, meaning, localization, register, clarity and linguistic consistency.
Studying prompt structure, instruction design, output control and evaluation-oriented prompting through practical examples.
Current areas of professional work and development.
Independent public projects created for professional development and AI evaluation practice.
Study notes based on Anthropic's public prompt engineering materials, rewritten and expanded with practical examples and an AI evaluation perspective.
View project →Original portfolio cases covering response comparison, factuality, instruction following and evaluator justification.
In progressOriginal evaluation examples focused on natural Turkish, localization, grammar, register and semantic accuracy.
Coming nextMy background in visual arts and creative work has strengthened my attention to detail, interpretation, visual judgment and human-centered evaluation.
Public work and professional profiles.