This study presents a hybrid approach for robust mass optimization of bio-capsules with a blunt-body geometry. The main objective is to minimize structural mass under uncertainties in material mechanical properties, including density, Young’s modulus, and Poisson’s ratio. The proposed framework is based on the All-at-Once (AAO) optimization framework, where in each iteration the optimal design point is determined using a genetic algorithm. The robustness of the design against variations in material properties is assessed through variance-based robust analysis. Latin Hypercube Sampling (LHS) is employed to generate statistical data and construct response surfaces for objective functions and constraints. Experimental tensile test data of aluminum specimens provide the basis for the analyses. Results indicate that the robust optimized design achieves about a 10% mass reduction compared to the baseline model while satisfying all design constraints with a two-sigma safety margin. Although the effect of uncertainties on mass is estimated to be less than 2%, a significant improvement in the buckling safety factor is observed. Flight analysis also demonstrates simultaneous improvements in aerodynamic performance (through reduced peak velocity) and flight stability (through increased final altitude). Overall, despite its relative simplicity, the proposed method achieves substantial enhancements in both the performance and safety of the capsule.