Taja Kuzman is a researcher at Department of Knowledge Tecnologies at the Jozef Stefan Institute« and at the company PC7 and is also a Phd student at International Postgraduate School of Jožef Stefan. She primarily works in the field of language resources, technologies, and computational linguistics, which means she prepares training data and develops or evaluates language technologies such as large language models like ChatGPT, as well as other models that are trained with machine learning to classify specific text characteristics, such as hate speech, text type, and topic.

Among other activities, she collaborates on projects at the Knowledge Technologies Department aimed at collecting large corpora of web, parliamentary, spoken, and other texts in Slovene and other South Slavic languages for linguistic research and the development of language models.

Within the infrastructure for language resources and technologies CLARIN.SI Taja, together with her colleague Dr. Nikola Ljubešić, leads the CLARIN knowledge center for South Slavic languages. CLASSLA,which develops and provides language resources and technologies, and trains researchers on their use. At PC7, she specializes in language models, developing and evaluating advanced solutions for services. PandaChat. These enable innovative information retrieval from documents through conversational chatbots and provide customer support on company websites via specialized chatbots.

Sobota, 25. 5., ob 14.00