The transfer of thermal energy is intrinsically associated with the generation of entropy. Here, we investigate the entropy generation number in nanofluid thin film flow over a radiated stretching sheet subject to different shapes of titanium oxide and copper nanoparticles in base fluid water. In addition, the impact of the magnetic field, viscous dissipation, and Joule heating were also considered. Using a suitable similarity transformation, the momentum and energy equations that describe the flow are converted into ordinary differential equations. These equations are then solved numerically using bvp4c in MATLAB. Graphs and explanations show how different physical parameters affect the velocity profile, temperature distribution, Nusselt number, skin friction coefficient, entropy generation number, and Bejan number. As the Brinkman number went up, fluid friction and Joule loss caused more entropy generation, but the Bejan number showed the opposite trend. In a certain situation, it is also important to note that as the value of the radiation parameter goes up, the irreversibility of heat transfer becomes more important. In addition, spherically shaped nanoparticles have the lowest entropy among all other forms. Hence, nanoparticles have the ability to reduce the amount of energy lost in any physical process and increase the overall productivity of systems. Understanding the flow and heat transmission within a thin film has many practical applications in industry, including aerodynamic lamination of plastic sheets, wire and fiber coating, food processing, annealing and thinning of copper wires, reactor fluidization, and many more.