Christos Kallidonis
Senior GPU Machine Learning Software Engineer at yasp
Germany
Summary
👤
Senior🎓
Top SchoolChristos Kallidonis is a Senior GPU Machine Learning Software Engineer with 11 years of experience bridging computational physics research and production-grade GPU software for medical imaging, quantum acceleration, and autonomous systems. He holds a PhD in Computational Sciences and has a proven track record of designing and implementing high-performance C++/CUDA solutions that scale on top supercomputers and industry GPUs, with applied impact on radiotherapy devices that improve patient outcomes. Equally fluent in Python-based ML stacks (PyTorch, TensorRT, TensorFlow, JAX) and HPC paradigms (MPI, OpenMP), he moves fluidly between research-grade algorithms and hardware-aware, production kernels. Colleagues describe him as a standards-setter who thrives under pressure and rapidly converts new concepts into reliable, optimized code. A less obvious strength is his history of turning once-intractable scientific problems—previously requiring petabytes and months of compute—into feasible workflows through bespoke algorithmic and parallelization strategies. Based in Germany, he currently develops cloud-native CUDA kernels and generative AI compiler components for scalable, hardware-agnostic model training and inference.
11 years of coding experience
8 years of employment as a software developer
Master's Degree (MSc) Physics, Master's Degree (MSc) Physics at University of CyprusDoctor of Philosophy (PhD) Computational Sciences, Doctor of Philosophy (PhD) Computational Sciences at The Cyprus Institute
Greek, English, Spanish, German
Github Skills (14)
petsc10
cuda10
hip10
opencl10
mpi10
lattice10
gpu10
multi-gpu10
infiniband9
nvidia-docker4
nvidia4
cudnn3
mass1
gpu-acceleration1
Programming languages (3)
Github contributions (5)
ckallidonis/mugiq
Dec 2019 - Feb 2021
Interface for MultiGrid deflation using QUDA on nVIDIA GPUs
Contributions:17 PRs, 116 pushes, 25 branches in 1 year 2 months
cudamultigridnvidia-dockernvidiagpu
ckallidonis/quda
Aug 2015 - Feb 2020
QUDA is a library for performing calculations in lattice QCD on GPUs.
Contributions:117 pushes, 84 branches in 4 years 6 months
cudaqcdgpu-accelerationmpicalculations
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.