About

I am a NeuroAI researcher currently finishing my PhD in Systems Neuroscience at Radboud University, supervised by Prof. Pascal Fries at the Max Planck Institute for Biological Cybernetics (Tübingen) and the Ernst Strüngmann Institute (Frankfurt). My thesis, Latent Structure in Brain and Behavior: From Geometric Brain State Decoding to Saccadic Rhythmicity, combined Riemannian geometry, causal deep learning, and stochastic modeling to decode brain states and characterize neural dynamics across species.

After my internship last year, I will join Meta in June 2026 as a Research Scientist in New York, where I will develop real-time deep learning models for peripheral neural interface decoding.

Contact Me

Research Interests

• Neural decoding & real-time models
• Deep learning for neuroscience
• Brain foundation models
• Multimodal Brain-computer interfaces (fNIRS, EEG, EMG)
• Riemannian geometry for neural data
• Saccade dynamics & visual neuroscience
• Sleep, memory & closed-loop neurostimulation

Featured Projects

SODEEP: Real-Time Slow Oscillation Detection

Causal CNN (~100k params) achieves F1 of 0.967 with 39 ms latency, replacing classical methods.

PyTorch Causal CNN Real-Time

Riemannian Geometry Boosts fNIRS Classification

Block diagonal SPD covariance kernels for fNIRS. 89% two-class and 70% eight-class accuracy.

SPD Manifolds pyRiemann fNIRS

fNIRS Awareness Detection in Disorders of Consciousness

Individualized Riemannian pipeline. 100% sensitivity and 89% specificity at bedside.

Clinical BCI Riemannian fNIRS

Riemannian Manifold Analysis of Infant Memory

SPD covariance matrices reveal convergence of voice representations after sleep in 3-month-olds.

pyRiemann Laplacian Eigenmaps EEG