diff --git a/python-package/basedosdados/core/base.py b/python-package/basedosdados/core/base.py
index 65b54db88..488044550 100644
--- a/python-package/basedosdados/core/base.py
+++ b/python-package/basedosdados/core/base.py
@@ -33,6 +33,7 @@ def __init__(
         self,
         config_path=".basedosdados",
         bucket_name=None,
+        billing_project_id=None,
         overwrite_cli_config=False,
     ):
         """
@@ -50,6 +51,10 @@ def __init__(
         self.config = self._load_config()
         self._config_log(config.verbose)
         self.bucket_name = bucket_name or self.config["bucket_name"]
+        self.billing_project_id = (
+            billing_project_id
+            or self.config["gcloud-projects"]["staging"]["name"]
+        )
         self.uri = f"gs://{self.bucket_name}" + "/staging/{dataset}/{table}/*"
         self._backend = Backend(self.config.get("api", {}).get("url", None))
 
diff --git a/python-package/basedosdados/upload/storage.py b/python-package/basedosdados/upload/storage.py
index 62d68d001..57f6f0e98 100644
--- a/python-package/basedosdados/upload/storage.py
+++ b/python-package/basedosdados/upload/storage.py
@@ -38,7 +38,9 @@ def __init__(self, dataset_id: str, table_id: str, **kwargs):
         """
         super().__init__(**kwargs)
 
-        self.bucket = self.client["storage_staging"].bucket(self.bucket_name)
+        self.bucket = self.client["storage_staging"].bucket(
+            self.bucket_name, user_project=self.billing_project_id
+        )
         self.dataset_id = dataset_id.replace("-", "_")
         self.table_id = table_id.replace("-", "_")
 
@@ -124,7 +126,9 @@ def init(self, replace: bool = False, very_sure: bool = False) -> None:
                 )
             self.bucket.delete(force=True)
 
-        self.client["storage_staging"].create_bucket(self.bucket)
+        self.client["storage_staging"].create_bucket(
+            self.bucket, user_project=self.billing_project_id
+        )
 
         for folder in ["staging/", "raw/"]:
             self.bucket.blob(folder).upload_from_string("")
@@ -399,7 +403,7 @@ def delete_table(
         if bucket_name is not None:
             table_blobs = list(
                 self.client["storage_staging"]
-                .bucket(f"{bucket_name}")
+                .bucket(f"{bucket_name}", user_project=self.billing_project_id)
                 .list_blobs(prefix=prefix)
             )
 
@@ -469,7 +473,7 @@ def copy_table(
 
         source_table_ref = list(
             self.client["storage_staging"]
-            .bucket(source_bucket_name)
+            .bucket(source_bucket_name, user_project=self.billing_project_id)
             .list_blobs(prefix=f"{mode}/{self.dataset_id}/{self.table_id}/")
         )
 
@@ -483,7 +487,7 @@ def copy_table(
 
         else:
             destination_bucket = self.client["storage_staging"].bucket(
-                destination_bucket_name
+                destination_bucket_name, user_project=self.billing_project_id
             )
 
         # Divides source_table_ref list for maximum batch request size
diff --git a/python-package/basedosdados/upload/table.py b/python-package/basedosdados/upload/table.py
index 963530491..d7cc73d28 100644
--- a/python-package/basedosdados/upload/table.py
+++ b/python-package/basedosdados/upload/table.py
@@ -274,8 +274,10 @@ def _parser_blobs_to_partition_dict(self) -> Optional[dict[Any, Any]]:
             return
         blobs = (
             self.client["storage_staging"]
-            .bucket(self.bucket_name)
-            .list_blobs(prefix=f"staging/{self.dataset_id}/{self.table_id}/")
+            .bucket(self.bucket_name, user_project=self.billing_project_id)
+            .list_blobs(
+                prefix=f"staging/{self.dataset_id}/{self.table_id}/",
+            )
         )
         partitions_dict = {}
         # only needs the first bloob
@@ -558,12 +560,12 @@ def create(
                 exists on your bucket:
                 * `raise`: Raises a Conflict exception
                 * `replace`: Replaces the table
-                * `pass`: Does nothing
+                * `pass`: Do nothing
             if_dataset_exists: Determines what to do if the dataset already
                 exists:
                 * `raise`: Raises a Conflict exception
                 * `replace`: Replaces the dataset
-                * `pass`: Does nothing
+                * `pass`: Do nothing
             dataset_is_public: Controls if the prod dataset is public or not. By
                 default, staging datasets like `dataset_id_staging` are not
                 public.
@@ -587,9 +589,12 @@ def create(
 
         if path is None:
             # Look if table data already exists at Storage
-            data = self.client["storage_staging"].list_blobs(
-                self.bucket_name,
-                prefix=f"staging/{self.dataset_id}/{self.table_id}",
+            data = (
+                self.client["storage_staging"]
+                .bucket(self.bucket_name, user_project=self.billing_project_id)
+                .list_blobs(
+                    prefix=f"staging/{self.dataset_id}/{self.table_id}",
+                )
             )
 
             # Raise: Cannot create table without external data