Marcel Veselý

Marcel’s expertise spans natural language processing, computer vision, and speech processing, with hands-on experience across a range of tasks including text classification, textual similarity, image segmentation and detection, and speech recognition. Over the past five years, the last three full-time, he has worked at the intersection of applied research and engineering, building and deploying AI solutions across both academic and industry settings.

His current research interests center on speech and audio models, particularly speech-to-text and text-to-speech systems for low-resource languages, as well as AI Agents and their practical applications. He is currently involved in ALFIE, a Horizon Europe project, where he focuses on developing an agent-based service that coordinates and delegates work across project components. He has also contributed to another Horizon Europe project, AI4Europe, where he designed and built an API service that exposes AI-powered search capabilities, combining semantic similarity search with agent-driven filtering.

Beyond these, Marcel has worked on a range of industry projects spanning conversational AI, product retrieval and ranking, and machine learning solutions for business process automation. He holds a Master’s degree in Computer Science from Faculty of Informatics and Information Technologies, Slovak University of Technology, graduating in 2023 with a thesis focused on minimizing domain shift in medical image data, work that followed his bachelor’s thesis on applying explainable AI methods to medical imaging. Outside of project work, he regularly leads workshops and seminars on deep learning, NLP and LLMs for external clients and events.