Centralized MLOps Platform: From R&D to Inference in a Single Loop
summary
A case study of building an internal MLOps platform used by 20+ teams, covering technology choices driven by business needs and ML developer requirements. It explains why Kubeflow was chosen as the platform foundation and describes the platform’s end-to-end flow from R&D to inference.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 5:47Business and team needs
- 11:09CFlow components
- 15:41Research and resource access
- 16:24Model training and pipelines
- 23:16Inference and monitoring
- 26:09Resource types and allocation
- 28:41Refinements and automation
- 30:29Implementation results
- 30:51Q&A
related talks
22mBuilding a MLOps Platform with Kubeflow on GKE
53mExperience with an end-to-end MLOps pipeline: Practical insights and 3D architectural perspectives
36mIs the MLOps tooling the same as the DevOps tooling?
14mMLOps: DevOps for AI with KubeFlow Pipelines
30mMLOps for Continuous Learning
33mBuilding Resilient (ML) Pipelines for MLOps
42mMLOps 2025: A Journey into the Past and the Future
30mEnd-to-End MLOps with MLflow and Kubeflow