Baby Sitting Your AI Models in Production? STOP!
summary
Claims a reproducible, low-touch MLOps workflow using the CNCF sandbox ModelPack and its directory spec ModelKit that bundles code, model weights, notebooks and metadata into a content-addressable OCI-compliant container image. Demonstrates kit init / kit build producing an image pushed to Docker Hub, pulling just training data or the full model from a registry, and integrating with GitOps (Argo), KServe/Kubeflow plus model signing and attestations; demo build and push works as shown.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 3:34DevOps to model compute
- 6:00Compute evolution and OCI artifacts
- 8:07KubeOps and model kits
- 12:45Packaging AI models with KitOps
- 17:38Closing & Q&A
related talks
23mBringing Container-Native Simplicity to AI/ML
41mModelPack: Packaging ML models as OCI artifacts made easy
27mBuilding Secure MLOps Pipelines With KitOps + Argo Workflows
05mModelpack: Standardizing The Packaging And Distribution Of...
05mProject Lightning Talk: ModelPack: Bringing the OCI Standard to AI/ML
32mBridging DevOps and MLOps: Unifying Pipelines with KitOps and GitOps
23mAccelerating Model Deployment with MLOps
24mDemo of Kubernetes at the Edge for AI Model Deployments