Machine learning methods build regression models to predict binding affinity with superior predictive performance compared with classical scoring functions in protein-ligand docking programs. This study focuses on a workflow to construct a neural network model to calculate the inhibition of cyclin-dependent kinase 9 (CDK9). It combines Jupyter Notebooks, Molegro Virtual Docker (MVD), and Molegro Data Modeller (MDM). MVD performs docking simulations and determines energy terms, ligand descriptors, and scoring function values based on pose coordinates. MDM builds regression models taking the features determined with MVD. The choice of CDK9 is due to its potential as a target for anticancer drugs. This approach integrates MVD-MDM with Jupyter Notebooks, allowing an integrated exploration of the scoring function space to find an adequate regression model. The necessary inputs are binding affinity data available in BindingDB and the atomic coordinates of the target found in the protein data bank. All CDK9 datasets and Jupyter Notebooks discussed in this work are available at GitHub: https://github.com/azevedolab/docking#readme .