Cadmium (Cd) as a toxic element that is widely present in water, soil, and air has important effects on human health, therefore proposing an accurate and selective method for detection of this element is of importance. In this article, by employing full atomistic molecular dynamics (MD) simulations and density functional theory dispersion corrected (DFT-D3) calculations, the effects of 6-mercaptonicotinic acid (MNA) and l-cysteine (CYS) on the stability of gold nanoparticles (AuNPs) and their sensitivity against Cd2+ were investigated. The obtained results indicate that pure AuNPs are not stable in water, while functionalized AuNPs with CYS and MNA groups have considerable stability without aggregation. In other words, the functional groups on the surface of AuNPs elevate their resistance against aggregation by an increase in the repulsive interactions between the gold nanoparticles. Moreover, functionalized AuNPs have considerable ability for selective detection of Cd2+ in the presence of different metal ions. Based on the MD simulation results, MNA-CYS-AuNPs (functionalized AuNPs with both functional groups) have the maximum sensitivity against Cd2+ in comparison with MNA-AuNPs and CYS-AuNPs due to the strong electrostatic interactions. DFT-D3 calculations reveal that the most probable interactions between the metal ions and functional groups are electrostatic, and Cd2+ can aggregate functionalized AuNPs due to strong electrostatic interactions with MNA and CYS groups. Moreover, charge transfer and donor–acceptor analyses show that molecular orbital interactions between the functional groups and Cd2+ can be considered as the driving force for AuNPs aggregation. A good agreement between the theoretical results and experimental data confirms the importance of the molecular modeling methods as a fast scientific protocol for designing new functionalized nanoparticles for application in different fields.