Summary
David Kubeša is a data scientist with 11 years of experience who builds production-grade LLM agents and applied ML systems, currently designing a custom Retrieval-Augmented Generation framework in AWS for ČSOB and integrating Azure models for targeted use cases. His doctoral work at Charles University focused on multilingual named entity recognition and linking at scale, using Spark and large knowledge bases, which informs his practical approach to information retrieval and conversational AI. Previously at O2 he applied classical and deep learning methods to customer lifecycle problems on big-data platforms, blending business impact with technical rigor. David bridges academic research and enterprise deployment, notable for advancing the Kate chatbot via Rasa Calm and for bringing research-grade entity linking techniques into real-world NLP pipelines.
12 years of coding experience
7 years of employment as a software developer
Master’s Degree, Artificial Intelligence, Master’s Degree, Artificial Intelligence at Faculty of Electrical Engineering, Czech Technical University in Prague
English, Czech, French