Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

Hello team. I am attempting to programmatically register my workflows through `remote.register_script(...)` but I keep getting a `TaskReferenceNotFound` error saying that workflows.s3_ingest.initialize_resources was not found. Similarly, when using `remote.register_workflow(...)` I get an error on execution time that module workflows was not found. At the moment, I am attempting to run these workflows from a monorepo to utilize a local library imported with a relative import in poetry. Do I have to register each task on its own before registering the workflow with `remote.register_script()` ?

Attempting to register the task before registering the workflow yielded no difference, and I can see the task in the ui. I'm guessing it has to do with the versions, but I am not sure how to move forward.

Same thing happens when attempting to register the example workflow from the dockerfile template when using `register_script()`. If I use `register_workflow()`, registration works, but when executing the workflow, I get the `ModuleNotFoundError` mentioned earlier.

are the workflows and the registration call present in the same file? could you move the workflows to an adjacent module and import them in the file where you're registering when using `register_workflow`?

For more context, I have managed to fast-register the workflows from the CLI with a custom image to include the library, and moved on to the actual content of the workflows. But I will need to circle back to the registration soon-ish.

image.png

```from flytekit.remote import FlyteRemote
from flytekit.configuration import Config, ImageConfig
from wf.test_json import jsons_iter_wf
from flytekit.configuration import SerializationSettings, FastSerializationSettings


remote = FlyteRemote(config=Config.auto(), default_project="flytesnacks", default_domain="development")

_, native_url = remote.fast_package(root=".")

flyte_workflow = remote.register_workflow(
    entity=jsons_iter_wf,
    serialization_settings=SerializationSettings(
        image_config=ImageConfig.auto(img_name="..."),
        project="flytesnacks",
        domain="development",
        version="v1",
        fast_serialization_settings=FastSerializationSettings(enabled=True, destination_dir=".", distribution_location=native_url),
    ),
)```
^^^ register.py

I haven't made any progress with this issue, but I want to thank you for the help. When I tried your solution I was still getting a `ModuleNotFoundError` when executing the workflow. It's probably an error I made when setting up the package that I'm not seeing. I will circle back to this as I said previously, but for now I'm registering from the CLI

<@U073ADW1YCB> Any success solving this? I'm facing the same issue, but only with flytekit version `&gt;v1.11`