Lightning Talk: Build a Reproducible ML Workflow with Kubeflow Pipelines
summary
Kubeflow Pipelines provides a reproducible, portable ML lifecycle running on Kubernetes and Docker, with a Python SDK that compiles code-defined workflows into a dashboard GUI. Example pipeline is a three-step download retrain redeploy loop that automatically retrains on new data and redeploys a served model with a REST endpoint. The system tracks every run, stores metrics and metadata, and supports on‑premise, cloud or multi‑cloud execution.
AI-written summary · about
chapters (9)
- 0:00Start
- 0:16Machine learning will be synonymous with software
- 1:08Lack of ML lifecycle management
- 2:44Dashboard
- 3:20ML Workloads Scale with Kubernetes / Istio
- 3:31Kubeflow Pipelines for ML workflows
- 3:50Define a pipeline
- 4:14View experiment run results
- 4:31Summary
related talks
35mOrganizing data engineering around Kubeflow
29mML Engineering Kubeflow
36mMachine learning pipeline Kubeflow, let's start
39mOperating Deep Learning Pipelines Anywhere Using Kubeflow
54mScaling Machine Learning Pipelines in Cloud
05mCI/CD Pipelines and Machine Learning
39mDeploy your ML Workloads on Kubernetes with Kubeflow
51mMachine Learning using Kubeflow and Kubernetes