• Approximate Computing
• Embedded Systems & Reconfigurable Architectures
• Energy-Efficient System Design
• Computer Vision & Machine Learning
• Autonomous Vehicles & Sensor Fusion
• Hardware Security

My research aims to improve how edge autonomy functions by developing new accelerator architectures, approximate computing techniques, and software/hardware co-designs. I focus on adaptive approximation methods that boost energy efficiency and fault tolerance without sacrificing accuracy. I also work on creating low-power accelerators and ML/AI frameworks for real-world platforms. I study new vulnerabilities in modern embedded systems, such as fault injections, and create effective, simple defenses to protect edge computation on devices with limited capacity. Looking forward, my goal is to build reliable approximate accelerators in embedded systems and investigate how ML/AI can help identify and address issues related to hardware efficiency, ultimately improving robustness and resilience at the edge.

• Approximate Computing – designing adaptive approximation techniques for resource-constrained edge computing systems.

• Edge Autonomy – building and testing energy-efficient accelerators and autonomy frameworks on real platforms.

• Hardware Resilience – modelling emerging hardware vulnerability and developing mitigation hardware architectures.