Kubernetes Custom Resources and GPU Scheduling
Extend the Kubernetes API with Custom Resource Definitions and schedule GPU-backed workloads.
Extend the Kubernetes API with Custom Resource Definitions and schedule GPU-backed workloads.
Expose applications with Kubernetes Services and add traffic management with the Istio service mesh.
Harden a Kubernetes cluster with pod security policies and fine-grained RBAC for users and permissions.
Control how Kubernetes schedules and manages workloads: Jobs, StatefulSets, taints/tolerations, affinity rules, health probes and resource limits.
Build a custom serving environment to run TensorFlow models with extra dependencies.
Explore how TensorFlow Serving handles different request content types for inference.
Design and proposal for a feature store to serve consistent features for training and serving.
Build a big-data pipeline stack combining Airflow, Spark and Kafka clusters with Docker.
Send and parse JSON prediction requests against a TensorFlow Serving model.
Package and serve a custom TensorFlow model signature with TensorFlow Serving.