Konstantinos Zampogiannis
Senior Computer Vision Engineer at NVIDIA
San Jose, California, United States
Summary
👤
Senior🎓
Top SchoolKonstantinos Zampogiannis is a Senior Computer Vision Engineer with 12 years of experience specializing in 3D machine perception, currently contributing to ground truth for autonomous vehicles at NVIDIA. He holds a Ph.D. in Computer Science from the University of Maryland and previously led computer vision research and engineering at Magic Leap, where he built scalable pipelines for high-quality digital twins and improved headset SLAM robustness. His academic work produced cilantro, a lean C++ point-cloud library, and includes novel methods for non-rigid, topology-aware registration and 6DoF object tracking—efforts reflected in performance-focused open-source contributions such as OpenMP parallelization and VoxelGrid optimizations. Based in San Jose, he combines deep research rigor with production-driven engineering, routinely turning advanced geometric ideas into optimized, real-world systems.
12 years of coding experience
10 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Computer Science, Doctor of Philosophy (Ph.D.) Computer Science at University of MarylandMaster of Engineering - MEng Electrical and Computer Engineering, Master of Engineering - MEng Electrical and Computer Engineering at National Technical University of Athens
English, Greek, French
Github Skills (11)
algorithm10
numerical-optimization10
code-optimization10
algorithms10
openmp10
c-language10
kmeans10
cprogramming-language10
optimisation10
optimization10
3d9
Programming languages (2)
Github contributions (5)
kzampog/cilantro
May 2017 - Jun 2022
A lean C++ library for working with point cloud data
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2 reviews, 779 commits, 13 PRs in 5 years 2 months
Contributions summary:Konstantinos's commits focus on optimizing the performance of the existing C++ point cloud library, particularly within the context of KMeans and ICP algorithms. They implemented OpenMP parallelization techniques to achieve speedups in the computationally intensive parts of these algorithms. Furthermore, they improved the VoxelGrid class, including the optimization of the building of the lookup tables, and fixed a bug related to the distance calculation within the KMeans class.
k-meanshierarchicalspectral-clusteringmdsiterative-closest-point
kzampog/ICPCUDA
Apr 2017 - May 2020
Contributions:7 commits, 7 pushes in 3 years 1 month
cudasuper-fasticpgpucompute
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.