My research is focused around hardware-efficient deep learning, spanning FPGA and reconfigurable computing architectures, low-precision arithmetic, neuromorphic computing, and bio-inspired learning algorithms for spiking neural networks.

Current interests include (but are not restricted to):

• Neuromorphic computing and event-driven learning algorithms (e.g., EventProp) for spiking neural networks
• Low-precision and alternative number systems (POSIT arithmetic) for accuracy-preserving, resource-efficient hardware
• FPGA-based acceleration of deep learning models, including Transformer and spiking neural network architectures
• Reconfigurable computing and High-Level Synthesis (HLS) for real-time, embedded AI systems
• Bio-inspired computation and brain-inspired learning as a route to efficient, low-power hardware
• Hardware security for embedded and edge AI systems