Work Experience

Neural Decoding & Machine Learning Research Intern

Integrated Neurotechnologies Lab at EPFL — Geneva, Switzerland
May. 2025 – Aug. 2025

• Engineered a lightweight deep learning model using PyTorch for on-chip EMG spasticity detection, achieving 93% classification accuracy and enabling deployment in a portable patient-assistive neuroprosthetic device.
• Optimized encoder architectures to boost EMG-based finger position decoding by 5% (86% accuracy), supporting real-time robotic prosthetic control while maintaining low computational cost for embedded integration.
• Adapted the decoding model to EEG, ECoG, and intracortical spiking datasets, achieving up to 85% motor decoding accuracy and demonstrating robust cross-signal reliability for brain–computer interface applications.

Brain-Computer Interface Researcher

WATOLINK — Waterloo, Canada
Sep. 2024 – Present

• Evaluated SSVEP and motor imagery paradigms using Neurosity Crown and g.tec Unicorn EEG headsets, benchmarking their performance for a reliable brain–computer interface that can control a drone.
• Built a PyTorch-based deep learning pipeline to decode EEG signals into movement intentions in real time, achieving 70% classification accuracy.

Spiking Neural Network Researcher

WAT.ai — Waterloo, Canada
Sep. 2024 – May. 2025

• Investigated the potential of Spiking Neural Networks for decoding intracortical spiking data into kinematic outputs, demonstrating its application for low-energy and biologically inspired brain–computer interface systems.
• Developed Python scripts for spike–kinematics correlation data analysis, enabling evaluation of neural firing patterns and providing insights for brain-computer interface model development.

Neurogaming Research Assistant

Dr. John Munoz at the University of Waterloo — Waterloo, Canada
May. 2024 – Dec. 2024

• Developed a 3D neuroadaptive Unity game integrated with the Muse EEG headset which can dynamically adjust gameplay difficulty based on EEG activity to enhance player engagement.
• Collected and analyzed EEG data from 15+ participants, identifying EEG patterns linked to player performance and difficulty levels, such as the engagement index, theta-alpha patterns, and beta-gamma patterns.

Neuroscience Research Assistant

Dr. Simon Chen Lab at the University of Ottawa — Ottawa, Canada May. 2023 – Aug. 2023

• Designed and executed an experimental study with neuroscientists, integrating pupil tracking and optogenetics to examine how noradrenergic modulation affects motor learning in autism-model mice.
• Applied transfer learning to train a convolutional neural network for mouse pupil size tracking, achieving mean pixel error of 1.4 px for high-precision behavioral measurement.
• Built real-time Python-based software integrating deep learning models to automatically trigger optogenetic stimulation, achieving low-latency performance (50 ms delay) on low-end hardware.