End-to-end AI Marketplace with Execution Engine on Kubernetes
summary
Announces an open catalog called machine learning exchange that registers datasets models pipelines components and notebooks, can run on Kubernetes/OpenShift and offers a hosted read‑only site plus a deployable GitHub stack. Pipelines execute with q flow pipelines (Kubernetes CRD) and tecton; models deploy via care serving; datasets download into PVCs and notebooks run as batch pipelines. Includes Project Codenet (millions of code samples), prebuilt model containers, and trusted‑AI pipelines using adversarial robustness and fairness checks; demos show full execution, lineage, logs and working inference endpoints.
AI-written summary · about
chapters (12)
- 0:00Start
- 1:18Center for Open-Source Data and Al Technologies -CC
- 2:27Pillars of Al Lifecycle - Datasets, Model
- 3:05and Pipelines
- 5:16Speeding up Al Lifecycle!
- 7:47Announcing: Machine Learning exchange MLX
- 11:04View, download, and execute Pipelines
- 12:00View, download, and execute Pipeline Components
- 15:01Library of prepackaged notebooks. Register your own notebooks
- 16:19MLX Katalog: Curated High Quality Content Notebooks
- 22:49Pipelines - Integration with Kubeflow Pipelines and Tekton
- 23:41Watson Studio Pipelines Open Beta Release
related talks
25mSecuring AI Pipelines: Real-World Attacks on Kubernetes-Based AI Infrastructure
39mOperating Deep Learning Pipelines Anywhere Using Kubeflow
33mDeploying and Scaling Spark ML and Tensorflow AI Models
44mEngaging the Kubeflow Community: Building an Enterprise-Ready AI/ML Platform
42mAI – Addressing the Challenges of Today’s Data Scientists
57mDevOps for Machine Learning
17mDevOps for Private AI: A Scalable Platform Using Kubeflow
37mEnd-to-End ML Pipelines with KubeFlow and TensorFlow Extended (TFX)