CI/CD Pipelines and Machine Learning
summary
Explains how Kubeflow turns ML setup into a repeatable two‑step flow that generates and applies manifests so clusters, GPU drivers, storage and ML apps (Jupyter, TensorBoard, custom controllers) get provisioned from code; shows using a git‑ops pattern where configs are pushed to a repo and Argo CD syncs them to the cluster, with the UI showing out‑of‑sync diffs and automated rollouts; demo uses kfctl style commands and ties into existing CD tooling.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:44Machine Learning on Kubernetes
- 2:15KF Cuddle
- 3:24CoopFlow
- 3:40GoCD
- 3:59GoCD UI
- 4:18Change manifest
- 4:40Outro
related talks
42mCI/CD for Machine Learning
36mMachine learning pipeline Kubeflow, let's start
54mScaling Machine Learning Pipelines in Cloud
05mLightning Talk: Build a Reproducible ML Workflow with Kubeflow Pipelines
51mMachine Learning using Kubeflow and Kubernetes
29mML Engineering Kubeflow
14mKubeflow: Where ML Meets Kubernetes
14mMLOps: DevOps for AI with KubeFlow Pipelines