Cosmin Eugen Oancea

Cosmin Eugen Oancea serves as an Associate Professor in the Programming Languages and Theory of Computation section at the Department of Computer Science (DIKU), University of Copenhagen, Faculty of Science. His academic profile shows active engagement in research, teaching, and collaboration within the computer science community.

Education:

• PhD in Computer Science from Western University (completed November 20, 2005)
• Dissertation title: "Parametric Polymorphism for Software Component Architectures and Optimizations"

Oancea's research spans multiple domains within computer systems, with primary focus on static, dynamic and hybrid techniques for automatic parallelization (optimizing compilers), programming language design and implementation, memory management, and computer algebra. His recent work demonstrates strong connections between functional programming, array languages, and high-performance computing, with applications in financial modeling, automatic differentiation, and time series analysis. He has made significant contributions to the Futhark programming language ecosystem, focusing on compiler optimizations and performance tuning for GPU architectures.

His publication record shows consistent output in top-tier venues including SC (International Conference for High Performance Computing), IFL (Symposium on Implementation and Application of Functional Languages), and ARRAY (Workshop on Libraries, Languages and Compilers for Array Programming), with a notable concentration of publications from 2021-2023. His research demonstrates interdisciplinary connections between computer science theory and practical applications in finance, environmental data analysis, and machine learning infrastructure.

Oancea maintains active international collaborations across multiple countries, as indicated by the network visualization of his research partnerships. His work bridges theoretical computer science with practical implementations, particularly in the domain of array programming and compiler optimization for parallel architectures.