Denisa-Andreea Constantinescu
Elle - She/her
Website: https://esl.epfl.ch
Expertise
Sustainable and energy-efficient computing; high-performance computing (HPC); sustainable AI and data centers; heterogeneous and reconfigurable computing; hardware–software co-design; CPU/GPU/FPGA/CGRA architectures; performance and energy optimization; carbon-aware computing; Edge AI and TinyML; scientific computing; confidential and privacy-preserving computing.
Mission
My research aims to make high-performance and AI computing systems more sustainable, energy-efficient, secure, and scalable. A central objective of my work is to translate sustainability from a high-level goal into measurable and actionable decisions across the computing stack, from hardware selection and workload execution to data-center operation and interaction with the energy grid.
Current Work
My current research is organized around three complementary directions:
Sustainable AI and HPC infrastructure
I investigate how AI and high-performance computing systems can reduce their energy use and environmental impact through life-cycle assessment, carbon-aware computing, grid-responsive operation, resource management, and sustainable infrastructure and procurement decisions.
I investigate how AI and high-performance computing systems can reduce their energy use and environmental impact through life-cycle assessment, carbon-aware computing, grid-responsive operation, resource management, and sustainable infrastructure and procurement decisions.
Energy-efficient heterogeneous computing
I develop hardware–software co-design methods for heterogeneous and reconfigurable platforms, including CPUs, GPUs, FPGAs, and CGRAs. My work explores how architecture, algorithms, and software can be co-optimized for performance, energy efficiency, and scalability.
I develop hardware–software co-design methods for heterogeneous and reconfigurable platforms, including CPUs, GPUs, FPGAs, and CGRAs. My work explores how architecture, algorithms, and software can be co-optimized for performance, energy efficiency, and scalability.
Privacy-preserving and confidential computing
I study approaches for secure and privacy-preserving distributed computing, including confidential computing and federated learning, particularly for applications involving sensitive scientific and health data.
I study approaches for secure and privacy-preserving distributed computing, including confidential computing and federated learning, particularly for applications involving sensitive scientific and health data.
My research is motivated by real-world, compute- and data-intensive applications, including radio astronomy, urban digital twins, environmental monitoring, biomedical wearables, and digital health.
Dr. Denisa-Andreea Constantinescu is a researcher in sustainable and energy-efficient computing at the intersection of high-performance computing, artificial intelligence, computer architecture, and distributed systems.
She is currently a Postdoctoral Researcher in the High Performance Computing Group at the University of Basel and a Visiting Researcher at the Embedded Systems Laboratory at EPFL. Her research develops cross-layer methods spanning algorithms, software, hardware, and computing infrastructure to improve performance, energy efficiency, scalability, sustainability, and security.
Her work focuses on sustainable AI and HPC data centers, carbon- and energy-aware computing, heterogeneous and reconfigurable architectures, hardware–software co-design, and confidential and privacy-preserving computing, with applications in scientific computing, radio astronomy, urban digital twins, environmental sensing, biomedical wearables, and digital health.
From 2022 to 2026, she conducted postdoctoral research at EPFL’s Embedded Systems Laboratory, contributing to sustainable computing, scientific computing, Edge AI, and heterogeneous systems. In 2018, she was a visiting researcher at Northeastern University in Boston, where she worked on high-performance computing and socially aware autonomous robot navigation.
She received the Intel oneAPI Innovator Award and the SCIE-ZONTA Award. She holds a PhD in Mechatronics from the University of Málaga, a Master’s degree in Computer Engineering from the University of Málaga, and a Bachelor’s degree in Automatics and Applied Informatics from the University Politehnica of Bucharest.