Recent publications

Jarosław Hryszko, Artur Michałek, Adam Roman, Application of large language models in medical interview training: a study with medical students, BMC Medical Education 26:586 (2026), 17
Ewa Woźny, Jarosław Hryszko, Adam Roman, Leveraging Large Language Models for Software Defect Detection, Euromicro Conference on Software Engineering and Advanced Applications [SEAA] LNCS 16082 (2026), 125–142
Adam Roman, Michał Mnich, Jarosław Hryszko, "To Measure Is to Know", but Not in Software Engineering. A Call for Operational Definitions of Code Metrics, Euromicro Conference on Software Engineering and Advanced Applications [SEAA] Lecture Notes in Computer Science (2026), 80-89
Jarosław Hryszko, Adam Roman, Quantum machine learning algorithms in software defect prediction for cybersecurity, (2025), "AI and quantum computing in unique synergy for cybersecurity", Wydawnictwo Politechniki Śląskiej
Adam Roman, Jarosław Hryszko, Rafał Brożek, Enhancing Software Defect Prediction: Exploring the Predictive Power of Two Data Flow Metrics, vol. Communications in Computer and Information Science, vol 2028 (2024), "Evaluation of Novel Approaches to Software Engineering. ENASE 2023", Springer

Interests

I research how to teach machines to find defects in code - before users, reviewers, or students on an exam do. My work focuses on software defect prediction, applying machine learning to quality assurance, and - more recently - investigating whether LLMs can detect what traditional tools cannot. The answer: sometimes yes, sometimes they hallucinate. Much like students during exams.

I lead projects such as FREGE (continuous static analysis of code stored in public repositories), URANUS (an experiment comparing different approaches to risk assessment in software development projects), and several others that are not in the public domain.

I co-author certification syllabi and exam questions in the areas of AI and software testing (including expert level). Member of the scientific committee of the Expert Cybersecurity Training Centre at the Ministry of Defence.

Check out my Mastodon!