Patrick Mccormack
Senior Data Scientist
Cambridge, Massachusetts, United States
Summary
👤
Senior🎓
Top SchoolPatrick Mccormack is a Senior Data Scientist in Cambridge, MA with a PhD in particle physics from UC Berkeley and six years of experience translating cutting-edge research into production ML and simulation tools. A former particle physicist and postdoc at MIT, he has a proven track record improving detector algorithms and reconstruction efficiencies, and deploying ML models into large-scale HEP software like CMS’s CMSSW (including ParticleNet ONNX/PyTorch integrations and deepMET work). At Fidelity he applies statistical rigor and physics-grade simulation expertise to production data problems while retaining a strong engineering focus on configuration, deployment, and reproducibility. He combines deep probabilistic and simulation experience with practical back-end development for Monte Carlo generation, making him adept at bridging research-grade methods and industry-scale systems. Notably, his work has repeatedly targeted high-density, high-pileup challenges—skills that transfer to domains requiring robust inference under noisy, real-world conditions.
6 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy - PhD, Elementary Particle Physics, 4.00/4.00, Doctor of Philosophy - PhD, Elementary Particle Physics, 4.00/4.00 at University of California, BerkeleyBachelor of Science - BS, Physics and Mathematics, 3.98/4.00, Bachelor of Science - BS, Physics and Mathematics, 3.98/4.00 at University of Notre Dame
English, German, French
Github Skills (18)
simulation10
cms10
c-language10
simulator10
simulations10
configuration-management10
machine-learning10
offline-web-app10
offline-app10
physics10
cprogramming-language10
monte-carlo10
pytorch9
onnx9
cpp8
Programming languages (5)
Github contributions (5)
cms-sw/genproductions
Aug 2021 - Nov 2021
Generator fragments for MC production
Role in this project:
Back-end Developer
Back-end DeveloperContributions:4 reviews, 14 commits, 6 PRs in 3 months
Contributions summary:Patrick primarily contributed to the generation of generator fragments for Monte Carlo (MC) production, specifically for QstarToQW decay processes. They modified configuration files to define particle masses, decay widths, and other parameters within the Pythia8 framework. The contributions involved adjusting parameters for different Qstar and W boson masses, and also included the creation of new configurations for different Pythia8 tunes, enhancing the simulation capabilities.
productionfragments
cms-sw/cmssw
Apr 2022 - Jun 2022
CMS Offline Software
Role in this project:ML Engineer
ML EngineerContributions:56 reviews, 22 commits, 6 PRs in 2 months
Contributions summary:Patrick primarily contributed to the implementation and integration of ParticleNet, a machine learning model, within the CMS offline software framework. Their work involved adding SONIC versions of ParticleNet, including both ONNX and PyTorch configuration files. Furthermore, the user modified model file paths, converted debug statements to logDebug, and fixed configuration files, suggesting a focus on deployment and optimization of the ML models within the specified infrastructure. They also added code related to deepMET models.
cmscernweb-appc-plus-plusbackbonejs
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.