Impact of AI on Software Engineering

Permanent URI for this collectionhttps://hdl.handle.net/10125/112561

Browse

Recent Submissions

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item type: Item ,
    Collaborative LLM Agents for C4 Software Architecture Design Automation
    (2026-01-06) Szczepanik, Kamil; Chudziak, Jarosław A
    Software architecture design is a fundamental part of creating every software system. Despite its importance, producing a C4 software architecture model — the preferred notation for such architecture — remains manual and time-consuming. We introduce an LLM-based multi-agent system that automates this task by simulating a dialogue between role-specific experts who analyze requirements and generate the Context, Container, and Component views of the C4 model. Quality is assessed with a hybrid evaluation framework: deterministic checks for structural and syntactic integrity and C4 rule consistency, plus semantic and qualitative scoring via an LLM-as-a-Judge approach. Tested on five canonical system briefs, the workflow demonstrates fast C4 model creation, sustains high compilation success, and delivers semantic fidelity. A comparison of four state-of-the-art LLMs shows different strengths relevant to architectural design. This study contributes to automated software architecture design and its evaluation methods.
  • Thumbnail Image
    Item type: Item ,
    Rethinking Programming Skills in the Age of Generative AI
    (2026-01-06) Bolici, Francesco; Crowston, Kevin; Varone, Alberto; Fudge, Michael
    What does it mean to be skilled in a world where machines can now write computer code? We explore how generative AI is not only accelerating productivity, but reshaping the very meaning of programming expertise. Adopting a relational perspective, we focus on three interdependent skills that define effective human–AI collaboration: task framing, prompt design, and output interpretation. Drawing on research in programming skills development and human–AI interaction, we trace the emergence of hybrid forms of competence that blend technical reasoning with contextual judgment, skills like strategic prompting, critical debugging, and situated problem framing. These signal a broader shift in programming: from producing code to coordinating AI-assisted problem solving, requiring new forms of cognitive effort and evaluative thinking. As AI becomes an active collaborator, the focus is moving away from writing code line-by-line toward orchestrating adaptive systems. This transformation has deep implications for how technical skills are learned, applied, and socially valued in AI-mediated environments.
  • Thumbnail Image
    Item type: Item ,
    Engineering prompts while navigating the tradeoffs of LLM integration in First Responder Software Application
    (2026-01-06) Paul, Sonipriya; Azmee, Abm Adnan; Thomas, Dominic
    This paper explores the implications and complications of incorporating LLM for software engineering or LLM4SE in conjunction with prompt engineering and vibe coding, a term coined by Andrej Karpathy, using tools such as Cursor and GitHub Copilot into the development lifecycle of an end-to-end software application. We present a case here in which we use these tools to implement a modular MVC architectural system with a SQLite-SQLAlchemy backend to support the import process, classification, and triage of large-scale case data. To support the large-volume data ingestion without compromising system stability, we implemented multithreaded chunked imports, which in turn enhanced UI responsiveness, minimizing memory overhead and reducing crashing risk. In this paper, we argue that although LLMs supported certain key logic components, in order to build a robust engineering pipeline, traditional software engineering practices had to co-exist, highlighting the importance of having an expert in the loop. We also examine the extent to which domain researchers, i.e., those without experience in software development, can leverage vibe coding and LLM-assisted workflows to build complete, production-level end-to-end software systems.
  • Thumbnail Image
    Item type: Item ,
    Introduction to the Minitrack on Impact of AI on Software Engineering
    (2026-01-06) Wittek, Stefan; Gesing, Sandra; Salhofer, Peter; Karl, Ryan
  • Thumbnail Image
    Item type: Item ,
    Prompt Engineering Patents in Software Development: Trends, Themes, and Future Directions
    (2026-01-06) Kassab, Mohamad
    This paper presents the first comprehensive patent landscape of prompt engineering in generative AI–driven software development, analyzing 2,511 patent families filed up until April-2025 from Lens.org. Combining reflexive thematic coding and LDA-based topic modeling, we develop a taxonomy of innovation themes—from AI-assisted code generation and automated debugging to test-case synthesis and prompt optimization frameworks. Our analysis reveals exponential growth in filings, led by industry giants, and uncovers gaps in areas such as legacy code modernization and developer trust. We highlight open challenges in verification, safety, and human–AI collaboration that must be addressed to realize robust AI-augmented development tools. The study provides insights and establishes a reproducible foundation for empirical validation and design of next-generation AI-assisted development environments.