The Good, the Bad, and the Missing Parts of Kubeflow
Talkat Kubeflow Summit Europe 2024 · on Tuesday, 2024-03-19· at 15:50· Pavilion 7 | Level 7.3 | Room S02·
summary
User survey shows top usage is pipelines, notebooks, and KServe and many users deploy from raw manifests, causing upgrade and install pain; pipelines v2 has partial feature parity and plans to merge Argo and Tekton backends into one code base. Roadmap targets Kubernetes 1.29, CNCF graduation, project-wide arm64 support, a model-registry alpha, LLM APIs for the training operator and KServe, Spark Operator donation, and asks for testers and concrete feedback on installs and docs.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 0:30Agenda
- 2:02Kubeflow Architecture
- 3:03User Survey
- 3:35The Good
- 4:46The Bad
- 6:50The Missing Parts
- 8:39Roadmap
- 11:57Expectations
- 15:22Questions
- 16:01Feedback
- 22:07Conclusion
related talks
18mKubeflow Ecosystem: Past, Present, Futu...
31mKubeflow 2026 State of the Union: Orchestrating the Full ML Lifecycle
21mIntroducing Kubeflow (w. Special Guests Tensorflow and Apache Spark)
44mEngaging the Kubeflow Community: Building an Enterprise-Ready AI/ML Platform
36mIntroducing Kubeflow (w. Special Guests Tensorflow and Apache Spark)
42mKubeflow From the End User’s Perspective: The Good, The Bad, and The Ugly
31mIntroduction to Using and Use Cases of KubeFlow
36mIntroducing Kubeflow (w. Special Guests Tensorflow and Apache Spark)