Get access to the AML job which runs the current Python code.

Question

Get access to the AML job which runs the current Python code.

Jaume Amores 25

Hello!

I would like to get access to the job running the current Python code. I know it is possible using mlflow if I know the run_id, as follows:

this_job = client.jobs.get(run_id)

How can I do it without knowing the run_id ? Ideally, without knowing any job details, so that Azure ML figures out which is the job running the code and retrieves a handle to it.

Thanks!

Jaume

Jaume Amores 25 Reputation points

2025-11-19T05:28:31.5233333+00:00

I've been answered by the AI bot that I can do that by using mlflow.last_active_run(). However, this returns null if mlflow.start_run() was not executed by the current pipeline component.

Answer accepted by question author

0 additional answers

Your answer

Jaume Amores 25 Reputation points

2025-11-19T05:28:31.5233333+00:00

I've been answered by the AI bot that I can do that by using mlflow.last_active_run(). However, this returns null if mlflow.start_run() was not executed by the current pipeline component.

Answer 1

Hi Jaume Amores,

Thanks for the question.

If you want to access the job that is currently running your pipeline, Azure ML doesn’t provide a direct “get current job” command during execution. However, you can still access the job information in a couple of easy ways.

1. Use environment variables inside the job When a job is running, Azure ML automatically injects variables like the run ID and job name. For example, in Python you can get them like this:

import os
print(os.getenv("AZUREML_RUN_ID"))
print(os.getenv("AZUREML_JOB_NAME"))

You can then use these values to query more job details using the MLClient.

2. Use the Azure ML SDK to fetch job details If you know the job name or run ID, you can retrieve full information:

job = ml_client.jobs.get(job_name)
print(job.status)

3. Check the portal You can always open Azure ML Studio → Jobs to see the run that executed your pipeline, including logs, outputs, and metadata.

Hope this helps! Let me know if you need any further assistance.

Thankyou!

Share via

Get access to the AML job which runs the current Python code.

0 additional answers

Your answer