kubeflow

Why we included this project

Kubeflow is the open-source platform most teams reach for when they want to run machine learning on Kubernetes, and it earns its place here because it addresses the step where many projects stall: moving ML work from a laptop into a shared, reproducible cluster. It bundles the components you would otherwise assemble yourself, including Jupyter-based notebooks for experimentation, distributed training for large models, Katib for hyperparameter tuning, a model registry, and pipelines that orchestrate everything from data prep to serving. Platform engineers and ML leads can pick individual subprojects or deploy the full community distribution as a self-hosted AI platform. The central dashboard and unified Python SDK let data scientists run workloads without becoming Kubernetes experts, which makes it a solid base for teams that want one ML infrastructure instead of a patchwork of tools.