Decoding
the brain.
My research is applied AI for the brain — neural decoding: learning to turn signals from neural activity into meaningful computational representations. It is less a single method than a moving set of problems, which is much of what draws me to it.
Applied AI for Neurotechnology
My current research is on image reconstruction from neural signals — recovering aspects of what a person perceives from recordings of the brain. The work spans the full pipeline, from neural data gathering and preprocessing through model development and evaluation, across several modalities including fMRI, fNIRS, and EEG.
Part of this work is currently confidential. More will be shared as it matures.
Neural decoding and multi-session alignment
My Bachelor's thesis investigates how machine-learning systems can extract semantic information from neural signals recorded during visual perception. It addresses a central difficulty in neural decoding: signals vary substantially across sessions, and I study whether alignment and calibration methods can build representations that generalize between them. The project combines signal processing, machine learning, experimental neuroscience, and reproducible benchmarking.