I am a PhD Candidate in Electrical and Electronic Engineering at University College Cork, working at the intersection of AI, sensors, and hardware. About what happens when machine learning leaves clean benchmarks and has to reason from real world signals on sensing interfaces and computing hardware.
Much of my work starts from the same questions: how can we make imperfect signals useful when the real world does not give us clean data, and how can we build the physical systems that will make it feasible? Before my PhD, I worked in the semiconductor industry as a Sensor Systems Engineer, which gave me a practical respect for the on-device computing side of AI. Models run on devices, consume power, wait on memory, and are highly interconnected to the sensors that collects the data. That is the space I am interested in: AI that remains useful when it meets the constraints of the real world.
June 2026: Article accepted in IEEE Sensors Journal on clinically meaningful phonocardiogram denoising for CHD detection.
June 2023: Coauthored work on whether a pre-trained neonatal EEG model can be used for seizure detection in pediatrics, as part of my master's research.
October 2022: Started working as a Sensor Systems Engineer at Qualcomm in Cork, Ireland.
January 2022: Started as a Research and Teaching Assistant at University College Cork.
July 2021: Published IGARSS work on machine learning with environmental remote sensing data for dengue risk modeling in Brazil, as part of my bachelor's research.
September 2020: Started as a Research Assistant at Universitat Politecnica de Catalunya in Barcelona.
Machine learning for signals collected from the physical world, including medical sigals PCG, ECG, EEG, or others like IMU, and remote sensing data. I'm especially interested in building models and systems that would have a positive impact on people's life, like certain disease detection, prevention mechanisms, human-machine interaction, etc.
I am interested in modern AI as a research field in itself. I want to work on approaches that improve how models learn, generalize, adapt, and make efficient use of data and compute.
I am interested in how models are transformed from research code into systems that can run efficiently on real hardware. Model acceleration, quantization, embedded deployment, and system-on-chip design, as well as the engineering questions behind them: where computation happens, how data moves through memory, and how latency and power are controlled, etc.
IEEE Sensors Journal, 2026
34th Irish Signals and Systems Conference (ISSC), 2023
From my master's degree research.
IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2021
From my bachelor's degree research.
