removing columns from data stream to bigquery datafow job

[Gokulavasan]

Hi ,
I have a dataflow streaming job running with a datastream to big query template , now my source data base has only 4 columns but after the dataflow job my final table has so many columns like _metadata_timestamp, _metadata_read_timestamp etc , I have tried to remove these columns with the help of  udf parameter but now i am not getting any data in the staging or final table. this is my udf

function process(inJson) {

    const obj = JSON.parse(inJson);

    // Concatenate two fields: column1 and column2

    obj.Name = obj.First_name +" "+ obj.Last_Name;

    // Optionally, you can remove the original fields if needed

    delete obj.First_name;

    delete obj.Last_Name;

    delete obj._metadata_timestamp;

    delete obj._metadata_read_timestamp;

    delete obj._metadata_source_timestamp;

    delete obj._metadata_read_method;

    delete obj._metadata_source_type;

    delete obj._metadata_deleted;

    delete obj._metadata_change_type;

    delete obj._metadata_uuid;

    delete obj._metadata_log_file;

    delete obj._metadata_log_position;

    return JSON.stringify(obj);

  }

how can i remove those columns in the dataflow job itself?