Currently, there does not exist a way to see newly inserted values in a live data source such as a hive table when using the data wrangler. There should be a way to invalidate the cache from the UI so that recent data will be read when querying sql sources.
WORK AROUND: If data is changed in the backing table, it can be seen by restarting kylo-services.
New data can be viewed by resampling the data.