Blazej Mrozinski

About me

I am a psychologist and psychometrician who builds products. I spent twenty years in research methodology learning how to measure things people care about but cannot easily define: traits, competencies, interests, change over time. Now I build products: systems that make decisions about people, jobs and pay more accurate and easier to defend, and software where the domain logic is the hard part, such as legal letters and health explanations.

The two halves feed each other. Measurement tells me what a product has to get right, and building tells me whether the measurement survives contact with real users.

Working with me

People come to me when they need to design an assessment, defend a pay structure, take a product from a fuzzy idea to working software, or rebuild how their team works with AI. I am most useful when the raw material is expert judgment, complicated rules or fuzzy human variables, and the task is to turn them into something that can be measured, built and shipped.

My role usually starts before any code: framing the problem, choosing what to measure and how, defining the logic and edge cases, and setting the criteria for whether it works. The work with me page shows how each kind of engagement runs.

What I do now

I work through three businesses I co-founded: Digital Savages for product building, AI transformation and growth, including our own products, Prawomat and Symptomatik; Gyfted for talent assessment; and legitnyhr for EU pay transparency. I also co-founded dots, a mentoring membership, and build custom clinical and research instruments through badania.org. Client work includes EPAM, LSEG, gr8.tech, HSE Ireland and Relativity’s Relevate academies.

Academic background

I have been at SWPS University since 2006 and am an Assistant Professor in its Department of Research Methodology in Psychology, where I directed the Experimental Psychology Laboratory from 2010 to 2024. I teach methodology, psychometrics, data analysis and R.

My PhD in psychology (social cognition, 2024) asked why some aspects of a person are easier to bring to mind than others, a question that matters directly for assessment and feedback design. My research on self-concept, intergroup behavior and research methods is published in journals including the Journal of Personality and Social Psychology and Self and Identity; the full list is on the publications page.

How I work

Systems first. Evidence always. Then build.

Structure, logic, assumptions and validation criteria are written down before implementation starts, so everyone knows what is being built and how we will know it works.

If you have a problem with fuzzy logic underneath it, get in touch. I will tell you honestly whether I can help.