Research Projects

Spintronic Magnetic Tunnel Junctions (MTJs) for Memory and Neuromorphic Computing

We investigate MTJ devices for spin transfer torque (STT), spin orbit torque (SOT)-driven switching including for memory and neuromorphic computing. Our work integrates simulations with magneto-transport characterization to establish clear links between materials, interfaces, switching dynamics, and device metrics.

Resistive Switching Devices for Non-Volatile Memory and In-Memory Computing 

We develop and study resistive switching devices based on thin-film structures, with emphasis on understanding and controlling diverse switching mechanisms. A parallel objective is to build compact and phenomenological models suitable for circuit-level evaluation and system co-design. 

Magnetic Platforms for Magnetic Sensing 

We explore magnetoelectric and quantum device concepts for next-generation magnetic sensors that target high sensitivity and low power operation. Current efforts focus on thin-film integration, reducing the noise floor, improving linearity, and developing scalable readout electronics and calibration methodologies. 

Device-Aware Neuromorphic Computing and Learning

We develop device-aware computing frameworks that incorporate realistic non-idealities from neuromorphic devices, including conductance states, stochasticity, nonlinearity, and drift. We assess accuracy–energy–robustness trade-offs and enable algorithm–hardware co-optimization for practical in-memory and edge-AI systems.