from education to academia
Nov 2024 - present
Newton International Research Fellow (in data science and AI), Royal Society | University of Surrey, United Kingdom
During this position I will be using statistical and ML/AI techniques to:
- i) simulate convection on multiple temporal and spatial scales
- ii) disentangle multiple periodic signals in time series data
- iii) build neural posterior density estimators for parameter inference from time series data.
Nov 2023 - Nov 2024
Data Science Research Fellow, Max Planck Institute for Astrophysics, Germany
- Partnered with statisticians to develop a regression-based (matrix calculation) algorithm for removing seasonal trends in time series data.
- Partnered with Caltech researchers to quantify stochastic variability in quasi-homogeneously sampled high-cadence time-series data.
- Led the development, calibration, validation, and publishing of an algorithm for the time-evolution of convective transport in 1-dimensional hydrodynamical codes based on 3D hydro simulations.
Aug 2023 - present
Group Leader, BlackGem Consortium, Global
- Partnered with statisticians to develop and publish an algorithm for detecting periodicity in heterogeneously sampled time series data using non-parametric kernel regression.
- Managed an international team of over 20 members, overseeing project timelines and resource allocation.
- Improved data acquisition efficiency by 33% through process optimization and resource management.
Jul 2021 - present
Visiting Research Professor, KU Leuven, Belgium
- Supervised PhD student Luc IJspeert on algorithm and software developemnt for parametric time series models. (Graduated September 2024)
- Designed and supervised an AI and Data Science MSc thesis in collaboration with the University of Cape Town to develop a machine learning algorithm for classifying star types based on time series behavior. (Graduated June 2021)
- Designed and supervised an interdisciplinary MSc thesis in Statistics and Astrophysics, applying Gaussian Process regression to jointly model stochastic and periodic signals in time series data. (Graduated June 2021)
Jun 2021 – Nov 2023
Radboud University, Nijmegen, The Netherlands
- Developed an analysis pipeline for time series, including database querying, data cleaning, and statistical analysis.
- Validated a machine learning pipeline to assess data quality.
- Built an algorithm to detect deviations in periodic signals and created a parametric model to predict this behavior in discrete time series observations.
Jun 2020 – Jun 2021
Data Science Postdoc, KU Leuven, Belgium
- Developed an algorithm to model periodic signals in non-stationary time series.
- Collaborated with statisticians to create a framework for deriving parameter inferences from high-dimensional time-evolution models.
- Partnered with researchers at the University of Oxford to build a probabilistic model forecasting deviations in the behavior of triple star systems.
Aug 2016 – May 2020
PhD Astrophysics, KU Leuven, Belgium
PhD in Astrophysics focusing on modelling periodic signals in astronomical time series data, and building statistical methods to optimize high-dimensional models.
Sep 2015 – Jun 2016
MSc Astrophysics, KU Leuven, Belgium
Magna Cum Laude: 81.59%
Oct 2010 – Dec 2014
BSc Astronomy & Astrophysics, Villanova University,USA
Minor degrees in physics and mathematics. Cum Laude: 3.73 / 4.0