Building and Evaluating
AI-Driven Measurement Systems

Pipeline thinking > model tweaking

Olga Kononykhina

Interdisciplinary research scientist and consultant working at the intersection of data, AI, and society.
I combine technical and social perspectives to understand how classification systems and AI tools behave in real-world use; and how data quality, design choices, and human behavior shape their outcomes.
Drawing on 15+ years of international research and applied work in Europe, Africa, the US across data, AI, and societal measurement.
PhD in Statistics (2026), MS Applied Mathematics (2009), BA Sociology (2007)

Areas of Expertise

Occupational Coding & AI-Supported NLP Classification

Applied research on AI-supported occupational coding, identifying pipeline design and data decision points that drive up to 30% accuracy loss —> Read more

AI Training & Capacity Building

AI-related work and a practical courses on AI tools and methods for researchers, educators, and medical professionals
—> Read more

Data Ethics, Statistics, and Responsible Data Use

Research and communication at the intersection of statistics, data ethics, and governance, with a focus on ethical decision-making in data-driven systems —> Read more

Program & Project Evaluation

Consulting practice on data driven program and project evaluation

Governance, culture, civil society measurement

Design of composite indicators and measurement frameworks for complex social concepts