The emergence of multidrug-resistant pathogens, such as methicillin-resistant Staphylococcus aureus (MRSA), poses a significant threat to the global public health. Streptomyces species have been recognized as a prolific source of bioactive secondary metabolites, including antimicrobial compounds. In this study, we aimed to optimize the production of anti-MRSA compounds by Streptomyces sp. AR05; a strain isolated from hydrocarbon-contaminated soil, using an integrated approach combining response surface methodology (RSM), artificial neural networks (ANN), and genetic algorithms (GA). The strain was identified through 16S rRNA gene sequencing and exhibited significant genetic similarity to Streptomyces kurssanovii and Streptomyces ostreogriseus. Using the Plackett-Burman design, the most important variables affecting the anti-MRSA activity were found to be peptone, CaCO₃, and pH. These factors were optimized using Box-Behnken design, while RSM and ANN were utilized for modeling the experimental data. The predicted accuracy of ANN model was higher than that of the RSM model, with lower values of mean absolute percentage error (MAPE) and root mean square error (RMSE). Sensitivity analysis of the ANN model identified peptone as the most influential factor, followed by pH and CaCO₃. The ANN model was further optimized using GA, and the optimized conditions (5.34 g/ l peptone, 1.54 g/ l CaCO₃, pH 6.07) were experimentally validated, resulting in a 48.87 % increase in anti-MRSA activity compared to the initial conditions. The developed RSM-ANN-GA approach demonstrated the potential for enhancing the production of valuable antibacterial compounds from Streptomyces species and contributed to the global efforts to combat antimicrobial resistance.