The present research work employs the artificial neural network (ANN) based deep learning (DL) approach to analyze the parametric estimation in double-diffusive convective flow in an inverted T-shaped porous enclosure. The study develops and numerically simulates a Darcy-extended Brinkmann-Forchheimer mathematical model using the penalty finite element method. Minimal training data are gathered across various flow parameters for DL model training. The DL-based technique efficiently approximates convective heat and mass transport across unknown flow parameter sets, significantly reducing computational costs. Analysis reveals that higher Rayleigh numbers prominently trigger convective phenomena, offering insights into realistic impacts of other flow parameters within porous media.