Denis Mitana

Denis Mitana is a Research Engineer at the Kempelen Institute of Intelligent Technologies (KInIT), where his work sits at the intersection of natural language processing, machine learning, and real-world AI engineering. His interest in language technology took shape during his studies at the Faculty of Informatics and Information Technologies, Slovak University of Technology in Bratislava (FIIT STU), where he completed both degrees with Magna Cum Laude distinction. His bachelor’s thesis explored speech-to-text conversion using neural networks, while his master’s thesis tackled misinformation detection with deep neural networks. Together, these reflect a consistent drive to apply AI to meaningful, language-centred problems. This academic foundation established his focus on core NLP challenges such as low-resource language processing, named entity recognition, morphological analysis, and large language model development, areas he has continued to build on throughout his career.

Prior to joining KInIT, Denis spent nearly four years as an NLP Engineer at the Slovak National Corpus, Ľ. Štúr Institute of Linguistics, Slovak Academy of Sciences, where he developed deep expertise in low-resource Slovak NLP. He is the author of the first publicly available Slovak Named Entity Recognition (NER) model trained on manually annotated data, as well as a Slovak model for sentence segmentation, tokenization, lemmatization, and morphological analysis for the spaCy framework, tools that continue to serve the Slovak NLP research community. Alongside his research work, he gained further engineering experience in time series forecasting for the finance and energy sectors. He is also the co-founder of Nanlab, a Bratislava-based AI company he co-built from the ground up, where he leads both the business and an R&D team spanning NLP, generative AI, agentic systems, retrieval-augmented generation, and data extraction.

At KInIT, Denis contributes his engineering and research expertise to projects with practical and societal impact. He participates as a research engineer in Dig4Health, a project focused on personalised digital health risk management that leverages large language models, data analysis, and recommendation methods to support people living with obesity and related health risks. He is also involved in the development of a Slovak large language model, addressing the challenges of building high-quality generative AI capabilities for a low-resource language, and contributes as a backend software engineer to SLAIH. Through this work, Denis bridges the gap between software engineering excellence and applied AI research.