cv

Research Scientist

Stanford University

Carried out research in high-energy astrophysics and machine learning, working with Professor Roger Blandford

Professor of Astrophysics

University of Sao Paulo

• Led teams of 5 to 10 data and computational scientists in ML, predictive modeling and simulation projects. I oversee our research goals, core strategies and data science roadmap.

NASA Postdoctoral Fellow

NASA Goddard Space Flight Center

• Lead developer for BCES—a Python regression package now used by hundreds of astronomers worldwide.
• Conducted feature engineering and regression analysis on satellite data, uncovering causal relationships in black hole power and radiation, which led to a new empirical law and a publication in Science.
• Employed techniques like partial correlation, multivariate regression, and bootstrapping on data from 35 galaxies, revealing significant links between galaxy properties and black hole feedback.

PhD

Universidade Federal do Rio Grande do Sul

Physics

NASA Postdoctoral Fellowship

NASA

Highly competitive annual award for new Ph.D. graduates to conduct cutting-edge research at NASA Centers.

Affiliated Member

Brazilian Academy of Sciences

Honor membership awarded to 0.1% of early career scientists in Brazil.

Miriani Pastoriza Prize

Brazilian Astronomical Society

This annual award honors one outstanding astronomy researcher from a Brazilian institution, representing less than 1% of early-to-mid career astronomers.

Emergence of hot corona and truncated disc in simulations of accreting stellar mass black holes

Monthly Notices of the Royal Astronomical Society

This paper showcases simulations of complex chaotic systems and analysing the resulting data to obtain insight.

Black hole weather forecasting with deep learning: a pilot study

Monthly Notices of the Royal Astronomical Society

AI-Driven Forecasting Algorithms using convolutional neural networks to accelerate simulations, showcasing techniques that can be tailored to hard-to-predict systems.

Deep learning Bayesian inference for low-luminosity active galactic nuclei spectra

Monthly Notices of the Royal Astronomical Society

This paper introduces a deep learning method to rapidly interpolate spectral calculations for black holes, enabling efficient Bayesian inference of many parameters.

On the relation between mini-halos and AGN feedback in clusters of galaxies

Monthly Notices of the Royal Astronomical Society

Example of using bootstrapping to robustly quantify prediction uncertainty.

Gamma-ray observations of low-luminosity active galactic nuclei

Monthly Notices of the Royal Astronomical Society

Here I use regression techniques for censored, sparsely sampled astronomical data.