How to create job id for Vertex AI manually or how to access job id in custom container?

Is there the possibility to manually specify the "Custom job ID" when starting a custom job on Vertex AI?

If not, is it possible to somehow access the automatically created job id from inside the custom container?

This is a late reply, but I'll put my answer to this question for future readers.

I was porting the custom jobs in the GCP AI platform to Vertex AI. Previously, AI platform was able to specify the job id when submitting the custom job like the below example. However, in vertex AI, the custom job id cannot be specified.

AI platform vs. Vertex AI (bash)

# AI platform
gcloud ai-platform jobs submit training ${JOB_ID} \
    --project=${PROJECT_ID} \
    --module-name trainer.task --package-path ./trainer \
    --region us-central1 --python-version 3.7 --runtime-version 2.11 \
    -- \        
    --param1=${param1} \
    --param2=${param2} \
    --param3=${param3}

# Vertex AI
gcloud ai custom-jobs create \
    --display-name=${DISPLAY_NAME} \  # cannot specify JOB_ID
    --region us-central1 \
    --project=${PROJECT_ID} \
    --python-package-uris='gs://[MODULE_PATH]/[MODULE_NAME]-0.1.tar.gz' \
    --worker-pool-spec=machine-type=e2-standard-4,replica-count=1,executor-image-uri='us-docker.pkg.dev/vertex-ai/training/tf-cpu.2-7:latest',python-module=trainer.task \
    --args=param1=${param1} \
    --args=param2=${param2} \
    --args=param3=${param3}

One possible solution to retrieve the job_id is to retrieve the job information with display_name filters.

Below is the python3 example of using ListCustomJobsRequest() and JobServiceClient().

Retrieve Vertex AI Job ID with filtering display_name tag (Python)

from google.cloud import aiplatform_v1  # extension

client = aiplatform_v1.JobServiceClient(
    credentials=credentials,
    client_options={"api_endpoint": "us-central1-aiplatform.googleapis.com"}
)
request = aiplatform_v1.ListCustomJobsRequest(
    parent=f"projects/{PROJECT_ID}/locations/{LOCATION_ID}",
    filter="display_name=[DISPLAY_NAME]",
)
page_result = client.list_custom_jobs(request=request)


for response in page_result:
    print(response)
    print(response.name)  # → "projects/[PROJECT_ID]/locations/[LOCATION_ID]/customJobs/[JOB_ID]"
    print(response.name.split('/')[5])
    print(response.state)  # → JOB_STATE_SUCCEEDED/JOB_STATE_PENDING

References

• gcloud ai custom-jobs create
• gcloud ai-platform jobs submit training
• list_custom_jobs()
• ListCustomJobsRequest()