kubeflow-pipelines

Machine Learning Pipelines for Kubeflow

Hybrid On-Prem DGX + GCP AI platform with Terraform-managed GKE, Workload Identity Federation, Artifact Registry, GPU self-hosted GitHub Actions runners, and agentic MLOps workflows.

🛡️ Simplify CI/CD with Pipeline Sentinel, automating failure analysis and predicting issues to keep your workflow smooth and efficient.

Kubeflow’s superfood for Data Scientists

Highly Scalable RecSys ML & Data system

Elyra extends JupyterLab with an AI centric approach.

ROCKs for Kubeflow Pipelines

Kubeflow Pipelines v2 customer-support intent router fine-tuned on BANKING77

Local rain forecast based on forecast and local weather station data

A Federated Learning framework for distributed Digital Twins on Kubernetes. Simulates diverse robotic environments to collaboratively train a robust RL policy using PyTorch, Temporal, and Kubeflow Pipelines, with Karmada for multi-cluster federation and MLflow experiment tracking.

Active Federated Learning: Target-Environment Probes, Active Weight aggregation, and Active Data (BC) fine-tuning. Demonstrated with PPO on CartPole, run locally or on Kubernetes via Kubeflow Pipelines, Temporal-orchestrated workers, and Karmada multi-cluster.

Prometheus metrics exporter for Kubeflow Pipelines (v2 / v2beta1) and Argo Workflows.

Automatically retrieve and process data from Public Health Ontario's (PHO) Water Testing Information System Electronic Notification (WTISEN)

Deploying a Multimodal Recommender System on Kubernetes featuring Cold Start handling, Bloom Filters, and Feature Caching.

Decision-grade comparison: Argo Workflows vs Kubeflow Pipelines for ML/MLOps pipelines on Kubernetes. Versions, licences, adoption figures and pricing verified against the GitHub API, CNCF project pages and vendor docs.

Common pipeline-editor components used in different clients (e.g. Elyra application, Web browser extensions, etc)

Intelligent contract automation platform - Vertex AI Pipelines orchestration, AI-powered document extraction, dual-source data reconciliation

3-stage Kubeflow Pipelines (KFP v2) ML pipeline on Vertex AI: BigQuery extraction → feature engineering → scikit-learn model training with full artifact lineage tracking. Components run in isolated Docker containers.

A curated list of awesome projects and resources related to Kubeflow (a CNCF incubating project)