You can just add resources and limits to that job so you can prevent it from being OOM killed. One thing to note is how resources and limits work. I have explained some scenarios in this answer on Stack. It is worthy to check the official documentation to avoid common mistakes and also CPU and Memory resources/limits work a little bit different.

About the default values of resources and limits it depends on where is your cluster. For example in my kubeadm I do not have any resources or limits by default in default namespace:

kubectl describe ns default
Name:         default
Labels:       <none>
Annotations:  <none>
Status:       Active

No resource quota.

No resource limits.

And this is for GKE:

So your choice is to set up Requests and Limits for your namespace or just add it into the Job spec, like in this example:

apiVersion: batch/v1beta1
kind: CronJob
metadata:
  name: testing-hello
spec:
  schedule: "*/1 * * * *"
  jobTemplate:
    spec:
      template:
        spec:
          containers:
          - name: hi-there
            image: busybox
            args:
            - /bin/sh
            - -c
            - date; echo "Hello, World!"
            resources:
              requests:
                cpu: "100Mi"
              limits:
                cpu: "100Mi"
          restartPolicy: OnFailure

Also usually you would just set resources and limits per namespace as it is described here.