Gold nanoparticles (AuNPs) are widely utilized in fields such as biomedicine, catalysis, electronics, sensors, optics, and cosmetics, due to their adjustable size and biological compatibility. The hydrodynamic diameter (HD) of AuNPs, which encompasses both the nanoparticle and its surrounding solvent layer, is a critical parameter influencing their stability, biological interactions, and overall functionality. This parameter plays a significant role in applications such as drug delivery, catalysis, and biosensing, as it impacts nanoparticle stability, cellular uptake, and optical characteristics. Experimental determination of the HD of AuNPs is often resource-intensive and challenging, requiring substantial optimization of synthesis parameters through iterative trial-and-error processes. To overcome these challenges and accelerate the synthesis of AuNPs for drug delivery applications, we introduce AuHD-Pred, a LightGBM-based classifier designed to predict AuNPs with HDs under 50 nm using 20 synthesis parameters. The parameters comprise both direct parameters—such as the concentrations and volumes of CTAC, ascorbic acid, gold precursor, and seed solution—and derived parameters, including flow rates and final concentrations, all of which play key roles in regulating nanoparticle nucleation, growth behavior, morphology, and stability. AuNPs with HDs of 50 nm or smaller are particularly effective for drug delivery. The model achieved an AUC-ROC of 0.89 \(_{\pm 0.01}\) and F1-score of 0.72 \(_{\pm 0.04}\) , highlighting the potential of machine learning in optimizing nanoparticle design and advancing the development of tailored nanomaterials.