Building Machine Learning Inference Through Knative Serverless...
summary
Claims a concrete serverless path for model inference by using KFServing on Knative to run scikit‑learn, TensorFlow and PyTorch models with YAML that defines min CPU/GPU, replicas and canary deployments. Shows explainability via Alibi Explain, and built-in pre/post processing and model swapping without manually wiring Kubernetes services, persistent volumes or custom schedulers. Presents small runnable demos (not executed live) and emphasizes reducing DevOps effort rather than inventing new model architectures.
AI-written summary · about
chapters (4)
related talks
39mIntroducing KFServing: Serverless Model Serving on Kubernetes
15mAI Model Inference on K8s
35mAdvanced Model Inferencing Leveraging KNative, Istio & Kubeflow Serving
28mExploring ML Model Serving with KServe (with fun drawings)
34mKServe in Production: Scaling Generative Inference the Cloud Native Way
40mServing Machine Learning Models at Scale Using KServing
30mAccelerate Federated Learning Model Deployment with KServe (KFServing)
26mMachine Learning Meets K8s