Fluids possessing advanced thermal capabilities are a requirement of today's world scientific technology and are an inherent vital part of diversified large-scale processes. As a result, the induction of nanometric-sized particles has been considered an emerging approach to achieve advanced liquids. Various combinations have been used to enhance the efficiency of nanofluids in thermal engineering systems. Recently, a suspension of three distinctively structured nanosized particles has been made in conventional liquid, termed as a ternary nanofluid. The current study primarily aims to optimize the design of a working system by inducing ternary nanoparticles composed of (GO), (Cu) and (Ag) in the flow of kerosene (base liquid) along elongated surface. The dimensionless version of the transport equations in the partial differential equations (PDEs) is regulated through pertinent similarity transformations into ordinary differential equations (ODEs). While providing a variable heat source, consideration is given to the radiation energy of secondary variations. The novel slip boundary constraint conceptualized by Thomas and Troian was assumed to be at the surface of the configuration. Computational simulations are executed by implementing shooting and Runge-Kutta (RK) procedures to obtain the results in a graphical and tabular manner. Subsequently, a machine-learning technique based on the Levenberg-Marquardt algorithm has been employed to predict the response of physical quantities to sundry parameters. The momentum profile was dominated by dispersing mono-nanoparticles compared to hybrid and ternary nanoparticles, whereas contrary aspects is attained for thermal distribution. The skin friction coefficient increased by up to 74% owing to the magnetic field factor. It is evident that the heat flux coefficient intensifies by up to 5% in the presence of quadratic thermal radiative energy compared to the linear radiative effect.