Vincenzo Innocente
Principal Software Scientist at CERN
Geneva, Geneva, Switzerland
Summary
👤
Senior
Top expert inHigh-Performance Machine Learning Computing
Vincenzo Innocente is a Principal Software Scientist at CERN with three decades of experience designing and optimizing simulation and reconstruction software for collider experiments, including long-term leadership roles in CMS as Chief Software Architect and head of the Performance Task Force. He has driven efforts to modernize HEP computing since 2008, leading multicore R&D and the Patatrack GPU initiative to port reconstruction algorithms and accelerate workloads with low-level CUDA optimizations. Vincenzo combines deep physics insight—having led tracking physics and detector-related projects—with hands-on systems expertise, notably implementing efficient prefix-scan and race-free CUDA kernels for CMS pixel workflows. A pioneer of machine learning in particle physics since 1990, he blends research rigor with production-grade software engineering to scale complex algorithms on heterogeneous platforms. Based in Geneva, he also brings cross-domain perspective from consulting for ESA’s Gaia ground software and a playful commitment to code and curiosity reflected in his GitHub motto: "code, & never grow up."
12 years of coding experience
14 years of employment as a software developer
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (11)
cuda10
c-language10
parallel-computing10
cprogramming-language10
performance-optimization10
data-structure9
datastructures-algorithms9
algorithm9
data-structures9
algorithms9
high-energy-physics8
Programming languages (8)
Github contributions (5)
cms-sw/cmssw
Jul 2013 - Jan 2023
CMS Offline Software
Role in this project:
Back-end Developer
Back-end DeveloperContributions:422 reviews, 2202 commits, 245 PRs in 9 years 8 months
Contributions summary:Vincenzo primarily focused on implementing and optimizing low-level CUDA kernels for processing pixel data. Their contributions involved addressing data races and improving the efficiency of prefix scan operations within the CUDA framework. Key tasks included porting existing pixel workflow and implementing the Cluster-to-TrackingParticle Associator to a new heterogeneous architecture for improved performance.
cmscernweb-appc-plus-plusbackbonejs
VinInn/ctest
Jul 2014 - Nov 2022
Contributions:946 commits, 911 pushes in 8 years 5 months
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.