MATLAB codes using the FT-based method, omitting the magnitude information, applying optimization, and AI methods are developed. The main code to use other subfunctions are "mainFunctionSingleBeamAtDifferentDirections.m", "mainFunctionSingleBeamAtDifferentDirectionsForOneDirection", and "mainFunctionMultiBeamAtDeterminedDirections.m". The first two m-files are developed to simulate a case, including just one main beam. However, the third one generates and simulates cases with more than one main beam. Note that for using the developed codes on your workstations/servers, you must change the paths in the code regarding your destination. This work has been funded by Fundaçao para a Ciencia e a Tecnologia (FCT), Portugal, under the Carnegie Mellon-Portugal Program (project ref. CMU/TIC/0080/2019) and by FCT/MCTES through national funds and when applicable co-funded by the EU funds under the project UIDB/50008/2020-UIDP/50008/2020. The codes and other materials developed under this project are distributed under the CC-BY-SA license, version 4.0 (https://creativecommons.org/licenses/by-sa/4.0/). Arrow3D code version 4.58 was obtained from MATLAB Central File Exchange, (C) Jeff Chang (2022) (https://www.mathworks.com/matlabcentral/fileexchange/1430-arrow3-version-4-58). Arrow3 version 4.58, https://www.mathworks.com. There is now an article related to this topic available in the repository entitled "Numerically Optimized Fourier Transform-Based Beamforming Accelerated by Neural Networks". If you wish to use these codes, please cite the article. For citation, you can use: "Kaboutari, K.; Abraray, A.; Maslovski, S. Numerically Optimized Fourier Transform-Based Beamforming Accelerated by Neural Networks. Appl. Sci. 2024, 14(7), 2866; https://doi.org/10.3390/app14072866." You can also find the article at "https://www.mdpi.com/2076-3417/14/7/2866".