I enjoy working at the intersection of computational neuroscience and neural engineering. I build data pipelines and analysis tools that turn messy, multi-channel neural and behavioral time-series into reproducible, queryable results.

ORCiD · LinkedIn

• eeg-feat-ext High-throughput pipeline for extracting cycle-level theta waveform features from human iEEG, built to support a waveform-shape control analysis which revealed HPC phase-amplitude coupling as a biomarker for memory.

• theta-feat-warehouse An Airflow pipeline that loads the eeg-feat-ext theta features into a DuckDB/SQL warehouse, gates them through 12 data-quality checks, runs paired permutation tests, and publishes the results to an offline dashboard and Tableau. Reads real iEEG through an NWB bridge (DANDI 000673).

• behavioral-data-discovery Visualizes open-field behavior with filtering and directional trajectory views. In the same place-cell study, I paired it with calcium-imaging analysis (CaImAn) of neural activity in rats.

Contributing to movement (UCL Neuroinformatics Unit), a Python library for animal-tracking data:

• Collective-behavior metrics: proposed a metric suite (#873), added compute_polarization() (#875), and a skeleton-agnostic anterior/posterior body-axis inference pipeline (#945).
• 3D vector utilities: Cartesian/cylindrical/spherical coordinate transforms (#948), being resubmitted as focused PRs per maintainer feedback (#1036, #1058).

• Neuronal encoding ⇄ decoding
• Signal processing · Time-series analysis · Data engineering · Statistical inference