In line with the global economic recovery, Indonesia's post-pandemic economic recovery is experiencing major challenge caused by uncertainty due to increasing inflationary pressure. Efforts to reduce inflationary fluctuations were carried out by the government in order to maintain optimal economic growth under the inflation targeting policy framework. Hence it is necessary to inspect the inflation threshold for economic growth in determining the inflation target. This study attempts to integrate inflation threshold estimation and nowcasting modelling of regional economic growth as an early warning system for achieving the government's economic growth target by utilizing several classical, machine learning, and ensemble models. Threshold estimation for the National level and East Java is applied to provincial groups obtained from clustering time series with dynamic time wrapping (DTW), author's justification, and the west-east region. In addition, estimates were also carried out in the period before Covid-19 as well as the entire period and the addition of predictor variables to test the sensitivity of the inflation threshold estimate. The results show that there was a non-linear relationship between inflation and economic growth with an inverted U shape. There are differences in the inflation threshold range due to provincial groupings, addition of control variables, both for the National level and for East Java. Evaluation of the inflation level in East Java and eight cities shows that in several periods, East Java inflation has exceeded Bank Indonesia's ITF target. Taking into account the estimated threshold range, it is recommended to set a more flexible inflation target at the regional economic level. Finally, the results of nowcasting with several specifications, such as the use of machine learning models and weighted ensemble models based on accuracy, as well as out-sample predictions with one-step ahead can improve the accuracy and consistency of the model for both short-term and long-term.