diff --git a/bigframes/session/__init__.py b/bigframes/session/__init__.py
index 3cb9d2bb68..c8ec877f2c 100644
--- a/bigframes/session/__init__.py
+++ b/bigframes/session/__init__.py
@@ -976,8 +976,7 @@ def read_pandas(
                   quota and your data cannot be embedded in SQL due to size or
                   data type limitations.
                 * "bigquery_write":
-                  [Preview] Use the BigQuery Storage Write API. This feature
-                  is in public preview.
+                  Use the BigQuery Storage Write API.
         Returns:
             An equivalent bigframes.pandas.(DataFrame/Series/Index) object
 
@@ -1026,7 +1025,7 @@ def _read_pandas(
         mem_usage = pandas_dataframe.memory_usage(deep=True).sum()
         if write_engine == "default":
             write_engine = (
-                "bigquery_load"
+                "bigquery_write"
                 if mem_usage > bigframes.constants.MAX_INLINE_BYTES
                 else "bigquery_inline"
             )
diff --git a/bigframes/session/bq_caching_executor.py b/bigframes/session/bq_caching_executor.py
index 736dbf7be1..dea6271bd9 100644
--- a/bigframes/session/bq_caching_executor.py
+++ b/bigframes/session/bq_caching_executor.py
@@ -594,7 +594,7 @@ def _upload_local_data(self, local_table: local_data.ManagedArrowTable):
         # Might be better as a queue and a worker thread
         with self._upload_lock:
             if local_table not in self.cache._uploaded_local_data:
-                uploaded = self.loader.load_data(
+                uploaded = self.loader.write_data(
                     local_table, bigframes.core.guid.generate_guid()
                 )
                 self.cache.cache_remote_replacement(local_table, uploaded)