生成式AI導入下軟體開發專案管理與工程師職能重塑：某軟體公司個案研究

Abstract

在生成式人工智慧（Generative AI）迅速發展並廣泛應用於軟體開發領域的背景下，軟體工程師的工作內容、角色定位與職能需求正經歷明顯的變化。本研究以質性研究方法為主軸，透過訪談八位不同資深程度的後端工程師與主管，運用工作分析觀點探討生成式AI導入後對軟體工程師日常工作流程、專案管理、人機協作與職能演化所帶來的實質影響。

本研究發現，初階工程師在生成式AI輔助下得以快速進行程式碼撰寫與除錯，然而也面臨過度依賴工具、缺乏基礎邏輯訓練的風險；資深工程師則展現出整合AI輸出並優化程式品質的能力，成為團隊中介者角色；技術領導與技術主管則更多投入於架構設計、流程優化與團隊知識管理，並善於引導AI作為知識延伸與決策輔助的工具。

此外，本研究亦發現AI工具的導入不僅重塑了個別工程師的工作內容，也促使團隊內部的專案協作邏輯產生轉變。例如：任務分配從「角色導向」轉向「能力導向」；工作流程從「線性分工」轉向「模組化交互」；人機協作型態由「工具式操作」演進為「共創式搭配」。

根據研究結果，本文整理出工程師職能演化動態地圖，揭示工程師於生成式AI環境下的角色重構與能力發展路徑，補足既有文獻中對軟體工程職場動態演變的描述不足，並對未來組織在人才招募、培育、專案管理與AI導入策略提供實務建議。

With the rapid development and adoption of generative artificial intelligence (Generative AI) in software development, the job content, role positioning, and competency requirements of software engineers are undergoing significant transformation. This study adopts a qualitative approach, conducting in-depth interviews with eight backend engineers at varying levels of seniority. Using a job analysis perspective, the research explores the impacts of Generative AI on engineers’ workflows, project management, human-AI collaboration, and competency evolution.

Findings show that junior engineers can accelerate coding and debugging with AI assistance, but risk over-reliance and lack foundational logical training. Senior engineers integrate AI outputs and improve code quality, often acting as team mediators. Technical leads and managers focus on system architecture, process optimization, and team knowledge management, leveraging AI for knowledge extension and decision support.

Moreover, AI tool adoption reshapes individual tasks and transforms team collaboration logic. For example, task allocation is shifting from role-based to competency-based models; workflows are evolving from linear to modular structures; and human-AI interactions are progressing from tool-based operation to co-creative collaboration.

Based on these findings, the study constructs a dynamic map of engineering competency evolution, illustrating how roles and skills are being redefined in AI-integrated environments. It also addresses gaps in current literature and provides recommendations for talent development, project management, and AI implementation.