Julius Von Kohout
Principal MLOps Architect at Kubeflow Freelancer and DHL Data & Analytics Lead MLOps Architect
Bonn, North Rhine-Westphalia, Germany
Summary
🤩
Rockstar🎓
Top SchoolJulius Von Kohout is a Principal MLOps Architect with seven years of hands-on experience designing and operating large-scale Kubeflow-based ML platforms for enterprises like DHL and Telekom. An elected member of the Kubeflow Steering Committee and active maintainer, he drives security, rootless container support, Ray integration, and CI/CD improvements across flagship repos such as kubeflow/manifests and kubeflow/pipelines. He has led the architecture of DHL’s company-wide ML/AI platform, consults on Kubeflow deployments, and mentors Google Summer of Code contributors, bridging open-source leadership with production delivery. A frequent KubeCon and Kubeflow Summit speaker and published blogger on hardening Kubeflow for enterprise use, he combines deep platform engineering with pragmatic security-first practices. Based in Bonn, Germany, he brings a rare mix of upstream open-source impact and hands-on platform execution that accelerates ML adoption at scale.
7 years of coding experience
8 years of employment as a software developer
Master of Science - MS, Informatik / Computer Science, Master of Science - MS, Informatik / Computer Science at The University of Bonn
German, English, German
Github Skills (18)
kubernetes10
docker10
kustomize10
kubeflow10
dockers10
cicd10
mlops10
kubernetes-pods10
istio10
ray10
mysql9
github-ci9
bash9
yaml9
go9
Programming languages (16)
Github contributions (5)
kubeflow/manifests
Mar 2021 - Jan 2023
A repository for Kustomize manifests
Role in this project:
DevOps Engineer & Infrastructure Engineer
DevOps Engineer & Infrastructure EngineerContributions:12 releases, 330 reviews, 13 commits in 1 year 10 months
Contributions summary:Julius primarily contributed to the integration of Kuberay and rootless container support within the Kubeflow manifests. This involved configuring Ray integration, modifying test scripts for rootless container testing, and updating Istio versions. Furthermore, the user managed the CI/CD pipelines, including updating Kustomize and other related tools.
kubeflowk8skuberneteskustomizemanifests
kubeflow/pipelines
Feb 2021 - Oct 2022
Machine Learning Pipelines for Kubeflow
Role in this project:DevOps Engineer
DevOps EngineerContributions:42 reviews, 17 commits, 39 PRs in 1 year 7 months
Contributions summary:Julius primarily focused on improving the Kubeflow Pipelines infrastructure. Their contributions include updating the MySQL deployment to version 5.7 and supporting its run as non-root, which enhances security. They also worked on relaxing the click version dependency in the SDK, along with addressing cache server issues by implementing maximum and default cache staleness settings, and ensuring the cache server integrates correctly with Kubeflow profile quotas.
pipelinetektondata-sciencemachine-learningmlops
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.