Ultrafine-grained (UFG) microstructure of Cu processed by large strain machining (LSM) is explored in order to create highly refined grain structures to achieve the highest strength while postponing the available nuclei for future recrystallization. The optimum solution is obtained theoretically using the Strength Pareto Evolutionary Algorithm (SPEA) and empirically using LSM. The thermal stability of the optimal solution is verified across the comparable LSM conditions using isothermal annealing curve. We also studied the kinetics of crystallization on the optimal solution using the Johnson–Mehl–Avrami–Kolmogorov (JMAK) theory. The optimal solution encountered leads to the latest time for the point where hardness start decline among a comparable sample conditions and lower the rate constant (1/τ) among LSM conditions.