sync partitioned dataset to GCS

Hello dataikers!

I have a flow which is like following: Postgres dataset -> sync to GCS -> sync to BQ

Everything is non-partitioned so far.

Dataset syncs to GCS pretty easily, but since size is pretty massive, while syncing from GCS to BigQuery

BigQuery throws error due to GCS single file limit -

Error while reading data, error message: Input CSV files are not splittable and at least one of the files is larger than the maximum allowed size. Size is: 16992300985. Max allowed size is: 4294967296

So, what I need to do is - write to GCS with partitions.

What I tried was:

gcs_dataset = dataiku_project.get_dataset(dataset_name)

gcs_def = gcs_dataset.get_definition()

gcs_def['partitioning'] = {
    'ignoreNonMatchingFile': False,
    'considerMissingRequestedPartitionsAsEmpty': False,
    'dimensions': [{'name': 'partition_id', 'type': 'value'}]
}
gcs_dataset.set_definition(gcs_def)

which kinda allows me to partition the dataset, but I couldn't find how to set partition dimensions afterwards.

And when I set them manually, I get following error:

Job failed: Root path does not exist: Root path of the dataset does not exist

So, my questions:

1. Is it possible to do what I want via API? (I don't need GUI solutions)

2. If yes, is it possible to do so without partitioning input dataset (Postgres)

3. If yes and yes, how to do so?

Thanks in advance!