This study employs a novel application of causal machine learning, specifically g-computation, to quantify aerosol effects on deep convective clouds (DCCs). Focusing on isolated DCCs in the Houston-Galveston region, we leverage comprehensive ground-based observations from the TRacking Aerosol Convection interactions ExpeRiment (TRACER) to estimate aerosol influences on convective core depth, intensity, and area. Our results reveal that greater aerosol number concentrations generally have a limited impact on convective core echo top height (ETH), with an increase of about 1 km (13 % of average ETH). This effect is observed under specific conditions, particularly when ultrafine particles are activated in updraft regions. Additionally, greater aerosol levels correspond to increased convective core intensity and area, though these changes remain within radar measurement uncertainties. In DCCs associated with sea breezes, aerosol effects are more pronounced, resulting in a 1.4 km deepening of ETH. However, this heightened effect could be attributed to the exclusion of key confounders such as boundary layer updrafts in the causal model. This study pioneers the application of causal machine learning to explore aerosol-convection interactions, shedding light on unraveling complex interplay between aerosols and meteorological variables.