Matthew Ragoza
Graduate Student Researcher
Dedham, Massachusetts, United States
Summary
👤
Senior🎓
Top SchoolMatthew Ragoza is a third-year PhD student in Intelligent Systems at the University of Pittsburgh whose research develops AI methods to augment scientific discovery, with applications in structure-based drug discovery and early disease detection from medical images. With 11 years of research and engineering experience across academia and industry (including an IBM Watson internship), he blends deep technical skills in machine learning, biomedical informatics, and full-stack development. He contributes to notable open-source projects like gnina (deep-learning molecular docking) and 3Dmol.js, improving data handling and molecular selection logic—work that reflects an emphasis on robust, scalable scientific tooling. Based in Dedham, MA, he pairs a multidisciplinary academic background (CS, Neuroscience, Chemistry) with practical software contributions that bridge model development and domain-specific data engineering.
11 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - PhD, Intelligent Systems, Doctor of Philosophy - PhD, Intelligent Systems at University of Pittsburgh
Github Skills (17)
molecule10
javascript10
convolutional-neural-networks10
caffe10
c-language10
machine-learning10
computational-chemistry10
graphic10
deep-learning10
cprogramming-language10
modeling10
webgl9
python9
cheminformatics9
data-structures8
Programming languages (6)
Github contributions (5)
gnina/gnina
Dec 2015 - Jun 2019
A deep learning framework for molecular docking
Role in this project:
Back-end Developer
Back-end DeveloperContributions:93 commits, 6 PRs, 33 pushes in 3 years 6 months
Contributions summary:Matthew made several commits focused on enhancing the deep learning framework for molecular docking. The primary contribution involved extending the data handling capabilities by adding a shape field to the Datum proto and implementing shape inference within the data transformer. Furthermore, the user merged updates, specifically concerning the batch normalization layer to incorporate support for generic ND-data. These modifications suggest a focus on expanding the framework's ability to handle diverse data formats.
cheminformaticsdockingdrug-discoverydeep-learningmolecular-docking
3dmol/3Dmol.js
Jan 2015 - Sep 2015
WebGL accelerated JavaScript molecular graphics library
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:102 commits, 19 PRs, 3 pushes in 7 months
Contributions summary:Matthew primarily worked on the 3Dmol.js library, contributing features related to molecular graphics. Their work involved modifying the existing setStyle() function to call selectedAtoms() and implementing a depth-first search algorithm for byres selection, demonstrating a focus on molecular selection logic. They also worked on testing the display and user interface of the library.
webglpdb-filesgraphics-libraryjavascriptaccelerated
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.