Dennis Fast

Dennis Fast is a Researcher affiliated with the Data Science and Text-based Information Systems (DATEXIS) research group at Beuth University of Applied Sciences Berlin. He joined the DATEXIS team as a research assistant in January 2022, collaborating on developing transformer-based biomedical language models with Paul Grundmann, Alexei Figueroa, and Conor Fallon. His work bridges computational engineering and data science, focusing on deep learning applications in natural language processing and physics simulations.

Education: MSc in Computational Engineering (specializing in Acoustic Simulations) from BHT Berlin (2018), and a second MSc in Data Science at BHT Berlin (started 2021). Professional experience includes roles as computational engineer and technical coordinator in European rail vehicle acoustics projects (Shift2Rail FINE1 & FINE2), where he specialized in data generation, analysis, and visualization.

Research interests emphasize deep learning algorithms, real-world applications of large language models, and interdisciplinary approaches combining physics simulations with machine learning. Active in acoustic modeling for transportation systems, contributing to noise prediction frameworks and railway innovation assessments.

Publications focus on acoustic analysis methodologies for railway systems, published in venues such as Euronoise and Internoise. Current projects explore biomedical NLP applications while maintaining expertise in simulation-driven engineering solutions.