Comparison of Traditional and AI-Mediated Approaches to Solving Engineering Problems

Burmistrov Aleksandr

Citation: Burmistrov Aleksandr, "Comparison of Traditional and AI-Mediated Approaches to Solving Engineering Problems", Universal Library of Innovative Research and Studies, Volume 03, Issue 01.

Copyright: This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

This article looks at how older engineering methods and newer AI-based ones differ when they are used to solve software-engineering problems. Most of the examples come from recent empirical studies, several review papers, and work on deep-learning models. The aim of the article is to look at how the older metric-based techniques compare with the newer ideas built around machine learning, deep learning, and large language models. The main result show that the traditional methods still help because they are clear and fairly steady to work with, but they do not capture the actual semantic behaviour of code very well. The AI models — especially the hybrid setups and the transformer ones — tend to score higher and can be used in more parts of the development process. These systems also introduce their own difficulties. Many of them depend on fairly large datasets, and it is often hard to work out what led the model to a specific output, even when the result looks reasonable. Re-running the same experiment does not always produce the same behaviour either, which can complicate evaluation. In practice, teams sometimes have to reshape parts of their workflow simply to make the tools usable. The article will interest practitioners seeking to streamline their daily work with the new technology.


Keywords: Traditional Engineering Methods, Software Defect Prediction, Deep Learning, Hybrid Models, Semantic Code Analysis, Large Language Models (LLMs), AI-Mediated Engineering, Software Quality, Machine Learning In Software Engineering, Engineering Workflow Transformation.

Download doi https://doi.org/10.70315/uloap.ulirs.2026.0301001