Choosing the technology stack for your application : what factors affect the decision making

Abstract

With the increasing complexity of artificial intelligence and modern software systems, the selection of software technology stacks has developed from a simple technical problem into a comprehensive decision-making issue involving performance, maintainability, compatibility and AI ecological adaptability. This thesis adopts a combination of narrative literature review and design science research (DSR) to analyze the key factors affecting the selection of software technology stacks and conducts practical verification through the construction of a Sloan Digital Sky Survey (SDSS) star spectrum classification web system. The research shows that technical factors, organizational factors and AI ecological consistency jointly affect technology stack decision-making. In the practical part, this thesis realizes a deep learning spectrum classification system based on technologies such as FastAPI, PyTorch, Astropy and JavaScript. The experimental results show that the system has good real-time performance and stability. The research further shows that in AI-driven software systems, a unified technology ecosystem can effectively reduce system complexity and improve development efficiency and maintenance capability. In the future, AI-assisted development tools will further affect the selection logic of software technology stacks.