Microbial cell factory development often faces bottlenecks after initial rounds of design-build-test-learn (DBTL) cycles as engineered producers respond unpredictably to further genetic modifications. Thus, deciphering metabolic flux and correcting bottlenecks are key components of DBTL cycles. Here, a 14-gene edited Pseudomonas putida KT2440 strain for heterologous indigoidine production was examined using both 13C-metabolic flux analysis (13C-MFA) and metabolite measurements. The results indicated the conservation of the cyclic Entner-Doudoroff (ED)-EMP pathway flux, downregulation of the TCA cycle and pyruvate shunt, and glyoxylate shunt activation. At the metabolite level, the CRISPR/dCpf1-interference mediated multiplex repression decreased gluconate/2-ketogluconate secretion and altered several intracellular TCA metabolite concentrations, leading to succinate overflow. Further strain engineering based on the metabolic knowledge first employed an optimal ribosome binding site (RBS) to achieve stronger product-substrate growth coupling (1.6-fold increase). Then, deletion strains were constructed using ssDNA recombineering. Of the five strains tested, deletion of the PHA operon (ΔphaAZC-IID) resulted in a 2.2-fold increase in growth phase production compared to the optimal RBS construct. After 72 h of batch cultivation, the ΔphaAZC-IID strain had 1.5-fold and 1.8-fold increases of indigoidine titer compared to the improved RBS construct and the original strain, respectively. Overall, the findings provided actionable DBTL targets as well as insights into physiological responses and flux buffering when new recombineering tools were used for engineering P. putida KT2440.