Johnu George
Kubeflow Steering Committee Member
Bengaluru, Karnataka, India
Summary
🤩
Rockstar🎓
Top SchoolJohnu George is an experienced DevOps and MLOps leader with 11 years building cloud-native ML infrastructure and production training/inference platforms from Kubernetes operators to CI/CD automation. As a Kubeflow Steering Committee member and Training & AutoML WG chair he maintains key projects like the PyTorch operator and Katib, and has repeatedly contributed to kubeflow/manifests to harden deployments and monitoring. He also led Nutanix’s GPT-in-a-Box initiative as Technical Director, blending systems architecture with scalable GenAI inference. An active MLCommons and Apache PMC contributor, he drives standards and benchmarking work (MedPerf) that connects research to clinical impact. Known for pragmatic refactoring and test-driven improvements across projects like Knative and KServe, he brings both deep backend/DevOps craft and governance experience to open-source ML platforms.
11 years of coding experience
10 years of employment as a software developer
Master of Science (MS) Computer Science, Master of Science (MS) Computer Science at Texas A&M UniversityBachelor's degree Computer Science, Bachelor's degree Computer Science at National Institute of Technology CalicutISC Maths, ISC Maths at St.Joseph's Public School Pattanakad
Stackoverflow
540reputation
74kreached
2answers
29questions
Github Skills (36)
kubernetes10
pytorch10
docker10
python10
webhooks10
testing10
kustomize10
configuration-management10
kubeflow10
machine-learning10
dockers10
cicd10
automation10
mlops10
go10
Programming languages (12)
Github contributions (5)
kubeflow/pytorch-operator
May 2018 - Jun 2021
PyTorch on Kubernetes
Role in this project:
DevOps Engineer
DevOps EngineerContributions:6 releases, 87 commits, 81 PRs in 3 years
Contributions summary:Johnu focused on improving the build and test processes for the PyTorch operator project. They implemented a minimal smoke test for distributed training using send and recv functions, built images for these tests, and integrated them into the existing run tests scripts. Additionally, they added a file for building Docker images during nightly releases, and made minor changes to fix build issues, contributing to the automation and reliability of the project's CI/CD pipeline.
pytorchkubernetesmachine-learning
kubeflow/katib
Dec 2018 - Jan 2023
Automated Machine Learning on Kubernetes
Role in this project:Backend & DevOps Engineer
Backend & DevOps EngineerContributions:3 releases, 94 reviews, 100 commits in 4 years 2 months
Contributions summary:Johnu primarily contributed to the Katib project by adding support for PyTorch jobs, including changes to the controller, metrics collector, and manifest parsing. They also updated the Kubernetes cluster version and made minor fixes to the deployment scripts. Furthermore, the user was involved in restructuring code for supporting v1alpha1 and v1alpha2 APIs and ensuring the correct labels are used.
xgboosthyperparametermachine-learninglightgbmautoml
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.