Application of Retrieval-Augmented Generation (RAG) Systems in Software Engineering Education: An Approach Based on Generative AI and DevOps

Software engineering education must be guided by developments in software engineering and aim for professional and educational development of students. This position paper explores the historical and conceptual evolution of software engineering as a discipline central to modern integrated design, software systems, and process science, tracing the trajectory from early structured models, through waterfall, to Agile and AI-augmented paradigms, and looks at how changes in application mix and modes of use, technological complexity, and business and enterprise needs have shaped development methodologies. Special attention is paid to the role of generative AI on the software engineering process and the role of the software engineer, altering workflows, collaboration, and enterprise architectures, and further enabling automation, digital transformation, and autonomous systems, shifting responsibilities and skill sets upwards. The article then considers corresponding shifts in software engineering education, highlighting the need for curricula that intertwine process thinking, systems theory, and ethical engagement, integrate with generative AI technologies, and place greater emphasis on “soft skills”, in courses that combine development of professional expertise with conceptual capabilities and an enhanced capacity for life-long learning. It then presents a set of requirements, options, and guidelines for software engineering education in the era of generative AI.

Learning Software Engineering principles, processes, methods and tools is a prerequisite to becoming a Software Engineer. The technological advancements and revolution in this new century are probing the boundaries of education and training in software engineering. This work tries to locate the shortcomings in existing approaches to Education in Software Engineering (ESE) in respect to today's technological advancements. We hereby proposed a new model for Education in Software Engineering to bridge the industry-academia gap.

Indian software industry has set up huge growth targets for future. These targets would be heavily affected by the software engineering (SE) education scenario in the country. The purpose of this paper is to provide a holistic understanding of SE education issues and challenges specific to Indian context, from both industry and academic perspective. This study is based on our (a) interaction with industry through SE education related projects, surveys and discussions. (B) Observations as an integral part of Indian SE educators' community. There is an urgent need for addressing these deep rooted issues, as the lack of proper SE education may be the single largest factor that may negatively affect the industry. Understanding these issues will help to identify the action items that initiate software engineering educational reforms in the country. We also discuss the essential and minimal set of SE knowledge, skills and dispositions that the Indian industry expects from engineers who will be joining the industry. Indian software industry is a global leader in software application development and maintenance services, but there very few studies on the issues and effects associated with SE education in the Indian context. Though this discussion is limited to India, but we believe that it represents the existing conditions in many developing countries where IT and ITES (IT enabled services) industry is gaining importance.

University Extension is an educational, cultural, and scientific process that establishes the exchange between academic and popular knowledge. Project-Based Learning (PjBL) consists of allowing students to confront the issues and problems of the world through a real project. PjBL has been adopted in Software Engineering (SE) teaching to motivate and engage students in studying and applying processes and methods. In this work, the university extension was integrated with the PjBL in teaching SE in a postgraduate course to bring students closer to real practices of software development, meeting a demand from the State of notorious value for society. Interviews were conducted with stakeholders, artifacts were delivered, and students received feedback from the professor and the stakeholders. Through a questionnaire, the students provided feedback about the practical activity, the learning method, and their perceptions regarding their evolution and dedication to group work. We can highlight the following findings: (i) the use of PjBL together with extension and the development of a real project was promising for the evolution of learning; (ii) greater motivation of the students, who evaluated: (iii) the learning method as satisfactory, and (iv) the practical activity and their evolution of learning as good and excellent. Our main contribution is the lessons learned that might guide future experiences that adopt PjBL in SE teaching and extension.

The article contains the results of a comparative analysis of digital infrastructure using the example of two universities of the Republic of Kazakhstan – KazNPU named after Abai and KSUTI named after Sh. Yessenov. The analysis examined key aspects of digital infrastructure, such as the availability and quality of Internet connections, the availability and use of educational platforms, digital libraries, and technical support and staff training. The article presents data obtained from surveys of students and teachers, as well as the results of monitoring the use of digital resources. The analysis is carried out as part of generalizing the experience of implementing modern online learning technologies, requirements for the organization of education, integration of cloud resources and online learning tools into learning processes at universities in the Republic of Kazakhstan. The results of empirical research were summarized based on a survey of teachers and students. This made it possible to identify the main trends in the use of digital technologies in the educational process of the analyzed universities.

Contribution: Results and conclusions on the use of transactive memory system (TMS) in software engineering (SE) education in student project teams that consist of undergraduate and postgraduate students are documented. This can be valuable for the adoption of this learning approach and team formation by other universities and institutes. Background: Preparing students for the SE industry requires bringing them in close collaboration to real-world tasks. This is usually performed via students carrying out software projects, working in teams. In such environments, the knowledge sharing and the interactions within the team are worth investigating, as they can affect the outcome of the student collaboration and can provide insight on how the collaboration is valued by the team members. Research Questions: 1) What is the effect of TMS on students in SE education? 2) Do parameters, such as gender, affect TMS development? 3) Does the inclusion of a project manager with more expertise affect TMS? Methodology: Students were divided into teams working on SE projects. Teams consist of both undergraduate and postgraduate students. Data from students on TMS for consecutive academic years were gathered. Findings: TMS is present in student teams and some of its scale items are affected by parameters, such as the team composition and the project type. TMS also affects performance.

This paper proposes a framework to support the development and use of tools that support the teaching of software and Web engineering. It describes the evolution of tools to support the teaching of software engineering from simple client-side tools that support drawing DFDs and UML diagrams to Web-based repositories that support process modelling. The paper specifically introduces the concept of and describes their use within the WWW. It also describes tools that automatically create automated walkthroughs for the World Wide Web (WWW) and their evaluation within a masters-level software engineering module. Conclusions are drawn about the general applicability of the framework and proposals are made for further work in this area.

In the Software Engineering Education (SEE) context, virtual worlds have been used in order to improve learning outcomes. However, there is a gap in the literature in order to characterize the use of the Metaverse for SEE. The objective of this work is to characterize the state of the art of virtual worlds in SEE and provide research opportunities and challenges to fill the limitations found. We conducted a systematic literature review, and we established 8 research questions that guided the study, as well as performed data extraction. We report on 17 primary studies that deal mostly with immersive experiences in SEE. The results show some limitations: few Software Engineering (SE) topics are covered; most applications simulate environments and do not explore new ways of viewing and interacting; there is no interoperability between virtual worlds; learning analysis techniques are not applied; and biometric data are not considered in the validations of the studies. Although there are virtual worlds for SEE, the results indicate the need to develop mechanisms in order to support the integration between virtual worlds. Therefore, based on the findings of the review, we established a set of components grouped by 5 layers to enable the Metaverse for SEE through fundamental requirements. We hope that this work can motivate promising research in order to foster immersive learning experiences in SE through the Metaverse.

This article articulates the nuanced challenges of integrating Generative AI (GenAI) into educational settings, aiming to dispel overly simplistic narratives driven by unwarranted enthusiasm or unfounded apprehensions. It introduces a conceptual framework for the application of GenAI within higher education, delineating four key strategies that leverage the dual roles of educators—as both creators and designers—while positioning GenAI as either a facilitative agent (creature) or a utilitarian tool. The identified strategies—Interactive Co-Creation, Adaptive Design, Learning Scaffold, and Efficient Structuring—underscore GenAI’s potential to revolutionize teaching methodologies by enabling personalized education, enhancing content quality, and expediting course development processes. Emphasizing GenAI’s capacity to cater to diverse student needs, simplify educational content creation, and foster engaging learning environments, the model provides educators with a roadmap for integrating GenAI into their instructional practices to harness the full potential of educational technology.Generative AI; educator’s role; instructional design

Aim/Purpose: This study investigated how undergraduate and graduate students in a teacher education program perceived, used, and ethically reflected upon generative AI tools. The problem is the unclear state of preservice teachers’ understanding and use of generative AI in educational settings, which this study aimed to explain. Background: Given generative AI’s growing presence in education, this study addressed a gap in understanding by examining preservice teachers’ perceptions, experiences, and ethical concerns, particularly in relation to self-regulated learning. Methodology: A convergent parallel mixed-methods design was used to gather quantitative and qualitative data from 73 students enrolled in a teacher education program at the University of Nevada, Las Vegas. Descriptive statistics, correlations, and thematic analysis were used for data analysis. Contribution: This study contributes to the emerging literature by providing insights into preservice teachers’ usage patterns of generative AI, the relationship between AI use and self-regulatory skills, and a detailed examination of ethical concerns, informing effective AI integration into teacher education programs. Findings: Text-based generative AI was the most familiar tool among participants. The primary reasons for the use included editing papers, searching for new ideas, and understanding course material. A significant positive correlation was identified between information search with the help of generative AI and resource management skill. Key themes identified by thematic analysis included perceived benefits for student engagement and creativity, AI’s potential as a writing assistant, and ethical concerns about plagiarism and misuse. Overall, preservice teachers demonstrated surface-level use and a general lack of in-depth knowledge about AI integration. Recommendations for Practitioners: Teacher educators should provide explicit training in generative AI, including practical guidelines and clear ethical frameworks to foster effective and responsible integration into classroom practice. Recommendation for Researchers: Future researchers should develop validated measurement instruments specifically designed for studying generative AI perceptions and usage and explore longitudinal impacts as preservice teachers transition into professional teaching roles. Impact on Society: By guiding preservice teachers toward informed and ethical use of generative AI, this study supports broader societal goals of promoting digital literacy, ethical technology integration, and enhanced learning experiences for future generations. Future Research: Subsequent research should further explore reasons behind the limited adoption of generative AI among preservice teachers, examine the long-term impacts of AI use on pedagogical practices, and investigate structured frameworks for promoting deeper AI integration within the teaching and learning domain.

Within the last few years constructivism has attracted both computer science and software engineer educators, because of its potential for improving academic learning. Currently, however, the application of constructivism is limited to examples in specific contexts so that it is difficult to transfer them to other institutions or even other courses. To apply effectively constructivism to software engineering and computer science education, instructors need a generic pedagogical model for translating constructivist principles into practice. This paper reports on the application of a pedagogical model for translating constructivism into a Web-based course in object-oriented software engineering.

ABSTRACTThe widespread applications of generative AI (GenAI) have sparked significant interest, with many organisations eager to leverage its transformative potential. Rather than focusing on individual organisations, this study examines GenAI integration within enterprise platforms, which are extensively adopted by many organisations and thus amplify both the benefits and risks of GenAI. We offer targeted recommendations for enterprise platform owners and their complementors, addressing challenges they face when integrating GenAI into these platforms. Drawing on a case study of Salesforce's experience, we recommend actions in three foundational areas – platform capability, architecture and governance – ensuring that our guidance is broadly applicable across enterprise platforms. In platform capability, we advise developing a unified GenAI stack built on existing platform services, offering generic and industry‐specific GenAI use cases to accelerate customer adoption and providing tools for customisation and creation of new use cases to enhance GenAI's transformational impact. For platform architecture, we recommend adding new layers for accommodating diverse GenAI foundation models and creating a trusted environment for secure data access, privacy and content monitoring. We also recommend implementing a prompt architecture to improve content relevance and accuracy. In platform governance, we recommend establishing new mechanisms to mitigate GenAI risks. Partnerships with GenAI providers and proactive investments in GenAI are essential to retain critical GenAI technologies. Personalised consultancy and training along with joint design and implementation with platform customers are also recommended. These combined actions, pursued in parallel across capability, architecture and governance, form a sustainable roadmap for GenAI integration in enterprise platforms.

Generative AI is revolutionizing software development by providing intelligent programming assistance, automating code generation, debugging, and refactoring. This paper evaluates the impact of AI-powered coding assistants, such as GitHub Copilot and OpenAI Codex, on developer productivity and code quality. We analyze AI’s effectiveness in improving coding efficiency, reducing errors, and maintaining best practices while addressing challenges such as bias, over-reliance, and security risks. Our study compares AI-assisted and human-written code using empirical data and controlled experiments across multiple programming languages. Key metrics, including defect rates, maintainability, and adherence to industry standards, are used to assess AI’s influence on code quality. Additionally, we explore its implications for software engineering workflows, education, and future advancements in AI-driven development. Findings reveal that AI-assisted coding enhances productivity but requires human oversight to mitigate risks like code vulnerabilities and inconsistencies. This research provides valuable insights into AI’s evolving role in software engineering, offering guidance for developers, researchers, and industry practitioners. DOI:https://doi.org/10.52783/pst.1668

Generative AI (GenAI), exemplified by Large Language Models (LLMs), such as OpenAI’s ChatGPT, is revolutionizing various fields. Central to this transformation is Data Center Networking (DCN), which not only provides the infrastructure support for GenAI operation, but also provisions GenAI services to users. Hence, this article explores the interplay between GenAI and DCNs, analyzing their symbiotic relationship and mutual advances. We begin by reviewing the current challenges of DCNs and GenAI-based solutions, such as data augmentation, process automation, and domain transfer. We then discuss the distinctive characteristics of GenAI workloads on DCNs, gaining insights that catalyze the evolution of DCNs to more effectively support GenAI. Moreover, to illustrate the seamless integration of GenAI with DCNs, we present a case study on GenAI-empowered DCN digital twins. Specifically, we employ an LLM equipped with retrieval augmented generation to formulate optimization problems for DCNs (e.g., resource allocation and routing) and adopt diffusion-deep reinforcement learning to solve optimization. The experimental results on a representative DCN optimization problem, i.e., knowledge placement, demonstrate the validity and efficiency of our proposals. We anticipate that this article can promote further research to enhance the virtuous interaction between GenAI and DCNs.

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Contribution:</i> This article contributes empirical insights on hybrid teaching of software engineering courses. Results include the systematic analysis of hybrid teaching attendance and interaction, perception of hybrid teaching, and grade distributions. Results are synthesized into eight evidence-based guidelines. <p xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <i>Background:</i> Hybrid teaching, i.e., teaching simultaneously to in-person and online students, is gaining an increasing adoption. However, how to improve the experience of students with respect to (w.r.t.) hybrid teaching is still an open question. <p xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <i>Research Questions (RQs):</i> How can the experience of students w.r.t. hybrid teaching be improved? <i>RQ</i> <inline-formula> <tex-math notation="LaTeX">$_{1}$</tex-math> </inline-formula>: Are there differences between in-person and online student attendance and interaction? <i>RQ</i> <inline-formula> <tex-math notation="LaTeX">$_{2}$</tex-math> </inline-formula>: What is the student perception of hybrid teaching? <i>RQ</i> <inline-formula> <tex-math notation="LaTeX">$_{3}$</tex-math> </inline-formula>: Is in-person and online supervision influencing grades of students? <p xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <i>Methodology:</i> A mixed-method empirical research process is used, by considering two Master courses in software engineering. The process leverages three data sources, namely, quantitative and qualitative data collected during lectures, a student survey, and student grades. Summary statistics, coding processes, and a statistical analysis are used to answer the RQs. <p xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <i>Findings:</i> Students prefer to attend more frequently online, as it provides (among other factors) flexibility and convenience, while coming at the cost of lower focus and interaction quality. Following in-person is statistically a better choice to gain a median grade, while following online can lead with more probability to a higher or lower grade. Various guidelines are presented, ranging from the hybrid classroom setup, to online student management, and course component design.