Authored the book 'Blocks: Towards Energy-efficient, Coarse-grained Reconfigurable Architectures' (Springer, 2021); German edition published in 2023
Published seminal paper 'How flexible is your computing system?' on defining and measuring flexibility in computing systems
Developed ConvFusion model for layer fusion in Convolutional Neural Networks to improve data reuse, energy efficiency, and performance
Delivered keynotes and tutorials at major conferences (e.g., DATE, HiPEAC, VLSISoC, DAC), including 'AI at the edge: Hype or Hope?' and 'Processors: From Intel 4004 to Neuro Chips'
Organized workshops such as NeuroEdge (brain-inspired computing), OSSMPIC (open-source massively parallel ICs), and FIRE on hardware accelerators
Research Experience
Conducting research at TU/e since 2001
Previously worked at Delft University of Technology (TUD) from 1986 to 2001
Leads the PARSE (Parallel Architecture Research in Eindhoven) lab
Leading large-scale programs such as the NWO-funded 'Efficient Deep Learning' program (launched Jan 2019, 7 sub-projects) and the 'ZERO perspectief' program on zero-energy systems (since early 2018, 5 sub-projects)
Involved in the EU HORIZON-CL4 project CONVOLVE (started Nov 2022) on smart edge processors with 18 partners
Principal investigator of the NWO project 'SNS: Self-healing Neuromorphic Systems' (kick-off Aug 2024) with partners including TUD, CWI, Innatera, IMEC NL, RUG, and TU/e