Engineering biological systems to test new pathway variants containing different enzyme homologs is laborious and time-consuming. To tackle this challenge, a strategy was developed for rapidly prototyping enzyme homologs by combining cellfree protein synthesis (CFPS) with split green fluorescent protein (GFP). This strategy featured two main advantages: (1) dozens of enzyme homologs were parallelly produced by CFPS within hours, and (2) the expression level and activity of each homolog was determined simultaneously by using the split GFP assay. As a model, this strategy was applied to optimize a 3-step pathway for nicotinamide mononucleotide (NMN) synthesis. Ten enzyme homologs from different organisms were selected for each step. Here, the most productive homolog of each step was identified within 24 h rather than weeks or months. Finally, the titer of NMN was increased to 1213 mg/L by improving physiochemical conditions, tuning enzyme ratios and cofactor concentrations, and decreasing the feedback inhibition, which was a more than 12-fold improvement over the initial setup. This strategy would provide a promising way to accelerate design-build-test cycles for metabolic engineering to improve the production of desired products.