This paper focused on optimizing deep drilling parameters based on Taguchi method for minimizing surface roughness. The experiments were conducted on CNC lathe machine using solid carbide cutting tool on material AISI 321 austenitic stainless steel. Four cutting parameters such as cutting fluid, speed, feed and hole-depth, each at three levels except the cutting fluid at two levels were considered. Taguchi L 18 orthogonal array was used as design of experiment. The signal-to-noise (S/N) ratio and the analysis of variance (ANOVA) was carried out to determine which machining parameter significantly affects the surface roughness and also the percentage contribution of individual parameters. Confirmation test was conducted to ensure validity of the test result. The results revealed that the combination of factors and their levels A 2 B 3 C 2 D 1 i.e. the machining done in the presence of cutting fluid, at a speed of 500 r.p.m. with a feed of 0.04 mm/s and hole-depth of 25 mm yielded the optimum i.e. minimum surface roughness. Further, the results of ANOVA indicated that all four cutting parameters significantly affected the surface roughness with maximum contribution from speed (27.02%), followed by cutting fluid (25.10%), feed (22.99%), and hole-depth (14.29%). It is also observed that the surface finish for deep drilling process can be improved effectively through this approach.