<p>This research studies&#xa0;flow dynamics for hybrid nanofluids at a stagnation point in porous media by adding the impacts of activation energy via a machine learning-integrated technique based on the Darcy-Forchheimer model and bio-convective transport. This work integrates ANN based machine learning techniques, including artificial neural networks and regression models, to improve predictive accuracy and computational efficiency. ANN model has data selection, network construction and evaluation of network performance by regression analysis. The trained ANN demonstrated excellent predictive capability, achieving a best validation performance with mean squared error of 2.84 × 10^–52.84 at epoch 4 while correlation coefficients reached R = 0.9977 (training), R = 0.9996 (validation) and R = 0.9984 (testing) confirming high accuracy and generalization. This work is focused on artificial neural network-based machine learning technique to investigate two-dimensional Williamson hybrid nanofluid flow over a stretching cylinder. The flow model is considered in a porous stretching cylinder using a hybrid nanofluid made of Molybdenum disulphide (MoS2) and Titanium oxide (TiO2) nanoparticles in ethylene glycol (EG) as a base fluid. TiO2 and MoS2 improve catalytic and adsorption properties, whereas MoS2 provides outstanding physical qualities and reduced friction. The governing equations of the model problem are transformed into ordinary differential equations using similarity variables and solved numerically. ANN based machine learning technique is capable to captures the interplay of activation energy, bio-convection, and porous media characteristics arising in industrial and biomedical applications, effectively. Moreover, thermal conductivity and flow characteristics of nanofluids for industrial applications are investigated. The present results are compared with results available in previous studies and main advantages of present work based on machine learning technique are elaborated explicitly.</p>