Cloud and Kubernetes for MLOps Engineers
Cloud and Kubernetes for ML is running training and serving on shared infrastructure. Kubernetes schedules containers with resource requests, GPU nodes and autoscaling, while Terraform describes cloud resources as code. This unit in the Serving and Infrastructure stage makes environments repeatable and costs controllable.
Stage: Serving and Infrastructure · Part of the MLOps Engineer roadmap · Last reviewed 19 November 2026
What to learn in Cloud and Kubernetes
- Deploy a model service on Kubernetes using a Deployment and a Service
- Set CPU, memory and GPU requests and limits for a pod
- Configure horizontal pod autoscaling on a metric
- Write Terraform to create a bucket and a cluster node pool
- Use spot or preemptible instances for interruptible training jobs
Cloud and Kubernetes interview questions
What are resource requests and limits in Kubernetes?
A request is the amount the scheduler reserves for a container when placing it on a node. A limit is the maximum it may use. Memory over its limit gets the container killed, while CPU over its limit is throttled.
Why use Terraform for ML infrastructure?
Terraform describes cloud resources in version-controlled files, so environments can be created, reviewed and reproduced the same way each time. Plan output shows changes before applying, and destroying unused infrastructure helps control cost.
Where Cloud and Kubernetes fits in the MLOps Engineer roadmap
Cloud and Kubernetes belongs to the Serving and Infrastructure stage (intermediate level), which takes about 6 hours. Low latency, high throughput.
Other units in this stage
Back to the full MLOps Engineer roadmap