Hybrid Chemo-Magnetorheological Finishing ( HC-MRF) process is an advanced surface finishing process that utilizes the interaction of Magnetorheological Finishing and chemical etching to reduce the surface roughness of the cemented carbide cutting tools. Herein, Murakami’s reagent is used as a Magnetorheological (MR) fluid carrier medium to trigger chemical etching and mechanical abrasion. The synergic action reduces surface finishing time and increases process efficiency. However, optimizing process parameters during the HC-MRF process is required to improve process efficiency further. Response Surface Methodology ( RSM) is used to develop the regression model to study the impact of process parameters (i.e. spindle speed, feed rate, and vol.% of Murakami’s reagent in MR fluid) on the surface quality. Three surface roughness parameters, namely average surface roughness ( Ra), skewness ( Rsk), and kurtosis ( Rk), are opted to analyze the surface characteristics of the polished surface. Furthermore, optimization is performed with the help of a Genetic Algorithm ( GA) to achieve the best surface quality following input parameters. The surface roughness ( Ra) is reduced by 88.97% as the initial Ra value decreases to 34.50 nm from 312.87 nm after the HC-MRF process. The results obtained from RSM and GA are in good agreement with the experimental results. The optimum value of feed rate, spindle speed, and vol.% of Murakami’s reagent in the MR fluid while achieving minimum surface roughness value (i.e. 31.26 nm) through GA are 1.2 mm/min, 23 rpm, and 53%, respectively. The measured kurtosis ( Rku) value of 0.64 at the optimum process parameter condition represents flat peaks on the surface roughness irregularities of the polished surface. Similarly, a skewness ( Rsk) value of 1.46 signifies that the number of peaks is higher than the number of valleys on the polished surface. Furthermore, it is observed that the Vol.% of Murakami’s reagent in MR fluid is the most significant process parameter compared with other parameters among all responses.