Simón Pedro González
Ph.D. Student in Computing (Artificial Intelligence) @ University of Utah, USA
Fulbright Scholar (2024–26) • M.S. in Computing (AI) (2026) @ University of Utah
B.S. in Computer Science (2021) @ National University of San Juan, Argentina
Contact:
simon.pedro.g@gmail.com •
GitHub •
LinkedIn
CV
About
- I’m a Ph.D. student in Computing (AI) at the University of Utah, advised by Jeff M. Phillips, working on algorithms and machine learning for scientific data.
- My current work focuses on ML-based detection and diagnosis of calibration errors in radio-interferometric data, in collaboration with Tanmoy Laskar (Physics & Astronomy).
- I’m also interested in multidimensional distribution distances, spatial scan statistics, anomaly detection, spatial data analysis, and scalable scientific computing.
- Previously, I developed open-source scientific software for astronomical data analysis, and worked for 5+ years as a software engineer and technical lead at Bridgenext (formerly Folcode).
Research
Accelerating Radio Interferometry Error Detection with AI
Machine-learning–based pipelines for the automated detection and diagnosis of calibration errors in radio-interferometric observations, combining simulated observations and machine learning.
Collaborators: Jeff M. Phillips (Kahlert School of Computing) • Tanmoy Laskar (Department of Physics & Astronomy), University of Utah.
My role: Research Assistant (Jan 2026 – Present).
Projects
SCLUDAM – Star CLUster Detection And Membership Estimation
SCLUDAM is a Python package for fetching data from GAIA catalogues, detecting star clusters, and estimating cluster membership probabilities.
Goal: Provide an accessible tool for detecting star clusters and estimating memberships from astrometric data.
My contribution: Author.
PyPI Package • Docs • Repo • Technical Paper • Featured in Alejo, González & González. (2024)
raptor-stats – Raptor Zonal Statistics Library
raptor-stats is a Python package for computing zonal statistics using Raptor (Raster + Vector) methods, which are more efficient at larger scales than standard raster methods such as masking. You can read more about the different approaches and benchmarks in this project report.
Goal: Provide fast zonal statistics methods for large raster and vector files.
My contribution: Author.
PyPI Package • Docs • Repo
pyscan-stats – Spatial Scan Statistics Library
pyscan-stats is a Python package based on the
pyscan wrapper by Michael Matheny,
providing fast C++ spatial scan statistics for anomaly detection. The algorithms
were developed by Jeff M. Phillips, Michael Matheny and collaborators.
Goal: Make these cool algorithms easier to use and build upon.
My contribution: Updated for new Python versions • Linux & Mac wheels & packaging • Cleaned and reorganized project • New features: fast grid constructor & area-limited scanning.
PyPI Package • Docs • Repo • Examples • Data Mining Class Project on Pollution Data Analysis.
Solar Explorer
A fun D3.js project exploring space missions and planetary data. Built for the 2025 Data Visualization course at the University of Utah — it ended up placing 2nd in the final project showcase!