arXiv subjects

Marcos Kalinowski

Publications and source records attributed to Marcos Kalinowski.

At least 37 records · Page 2Linked to original sources

arXiv2025

LLM-Assisted Thematic Analysis: Opportunities, Limitations, and Recommendations

[Context] Large Language Models (LLMs) are increasingly used to assist qualitative research in Software Engineering (SE), yet the methodological implications of this usage remain underexplored. Their integration into interpretive processes such as thematic analysis raises fundamental questions about rigor, transparency, and researcher agency. [Objective] This study investigates how experienced SE researchers conceptualize the opportunities, risks, and methodological implications of integrating LLMs into thematic analysis. [Method] A reflective workshop with 25 ISERN researchers guided participants through structured discussions of LLM-assisted open coding, theme generation, and theme reviewing, using color-coded canvases to document perceived opportunities, limitations, and recommendations. [Results] Participants recognized potential efficiency and scalability gains, but highlighted risks related to bias, contextual loss, reproducibility, and the rapid evolution of LLMs. They also emphasized the need for prompting literacy and continuous human oversight. [Conclusion] Findings portray LLMs as tools that can support, but not substitute, interpretive analysis. The study contributes to ongoing community reflections on how LLMs can responsibly enhance qualitative research in SE.

cs.SE↗

arXiv2025

DIP-AI: A Discovery Framework for AI Innovation Projects

Despite the increasing development of Artificial Intelligence (AI) systems, Requirements Engineering (RE) activities face challenges in this new data-intensive paradigm. We identified a lack of support for problem discovery within AI innovation projects. To address this, we propose and evaluate DIP-AI, a discovery framework tailored to guide early-stage exploration in such initiatives. Based on a literature review, our solution proposal combines elements of ISO 12207, 5338, and Design Thinking to support the discovery of AI innovation projects, aiming at promoting higher quality deliveries and stakeholder satisfaction. We evaluated DIP-AI in an industry-academia collaboration (IAC) case study of an AI innovation project, in which participants applied DIP-AI to the discovery phase in practice and provided their perceptions about the approach's problem discovery capability, acceptance, and suggestions. The results indicate that DIP-AI is relevant and useful, particularly in facilitating problem discovery in AI projects. This research contributes to academia by sharing DIP-AI as a framework for AI problem discovery. For industry, we discuss the use of this framework in a real IAC program that develops AI innovation projects.

cs.SE↗

arXiv2025

Investigating the Use of LLMs for Evidence Briefings Generation in Software Engineering

[Context] An evidence briefing is a concise and objective transfer medium that can present the main findings of a study to software engineers in the industry. Although practitioners and researchers have deemed Evidence Briefings useful, their production requires manual labor, which may be a significant challenge to their broad adoption. [Goal] The goal of this registered report is to describe an experimental protocol for evaluating LLM-generated evidence briefings for secondary studies in terms of content fidelity, ease of understanding, and usefulness, as perceived by researchers and practitioners, compared to human-made briefings. [Method] We developed an RAG-based LLM tool to generate evidence briefings. We used the tool to automatically generate two evidence briefings that had been manually generated in previous research efforts. We designed a controlled experiment to evaluate how the LLM-generated briefings compare to the human-made ones regarding perceived content fidelity, ease of understanding, and usefulness. [Results] To be reported after the experimental trials. [Conclusion] Depending on the experiment results.

cs.SE↗

arXiv2025

Towards a Framework for Operationalizing the Specification of Trustworthy AI Requirements

Growing concerns around the trustworthiness of AI-enabled systems highlight the role of requirements engineering (RE) in addressing emergent, context-dependent properties that are difficult to specify without structured approaches. In this short vision paper, we propose the integration of two complementary approaches: AMDiRE, an artefact-based approach for RE, and PerSpecML, a perspective-based method designed to support the elicitation, analysis, and specification of machine learning (ML)-enabled systems. AMDiRE provides a structured, artefact-centric, process-agnostic methodology and templates that promote consistency and traceability in the results; however, it is primarily oriented toward deterministic systems. PerSpecML, in turn, introduces multi-perspective guidance to uncover concerns arising from the data-driven and non-deterministic behavior of ML-enabled systems. We envision a pathway to operationalize trustworthiness-related requirements, bridging stakeholder-driven concerns and structured artefact models. We conclude by outlining key research directions and open challenges to be discussed with the RE community.

cs.SE↗

arXiv2025

Define-ML: An Approach to Ideate Machine Learning-Enabled Systems

[Context] The increasing adoption of machine learning (ML) in software systems demands specialized ideation approaches that address ML-specific challenges, including data dependencies, technical feasibility, and alignment between business objectives and probabilistic system behavior. Traditional ideation methods like Lean Inception lack structured support for these ML considerations, which can result in misaligned product visions and unrealistic expectations. [Goal] This paper presents Define-ML, a framework that extends Lean Inception with tailored activities - Data Source Mapping, Feature-to-Data Source Mapping, and ML Mapping - to systematically integrate data and technical constraints into early-stage ML product ideation. [Method] We developed and validated Define-ML following the Technology Transfer Model, conducting both static validation (with a toy problem) and dynamic validation (in a real-world industrial case study). The analysis combined quantitative surveys with qualitative feedback, assessing utility, ease of use, and intent of adoption. [Results] Participants found Define-ML effective for clarifying data concerns, aligning ML capabilities with business goals, and fostering cross-functional collaboration. The approach's structured activities reduced ideation ambiguity, though some noted a learning curve for ML-specific components, which can be mitigated by expert facilitation. All participants expressed the intention to adopt Define-ML. [Conclusion] Define-ML provides an openly available, validated approach for ML product ideation, building on Lean Inception's agility while aligning features with available data and increasing awareness of technical feasibility.

cs.SE↗

arXiv2025

Domain Knowledge in Requirements Engineering: A Systematic Mapping Study

[Context] Domain knowledge is recognized as a key component for the success of Requirements Engineering (RE), as it provides the conceptual support needed to understand the system context, ensure alignment with stakeholder needs, and reduce ambiguity in requirements specification. Despite its relevance, the scientific literature still lacks a systematic consolidation of how domain knowledge can be effectively used and operationalized in RE. [Goal] This paper addresses this gap by offering a comprehensive overview of existing contributions, including methods, techniques, and tools to incorporate domain knowledge into RE practices. [Method] We conducted a systematic mapping study using a hybrid search strategy that combines database searches with iterative backward and forward snowballing. [Results] In total, we found 75 papers that met our inclusion criteria. The analysis highlights the main types of requirements addressed, the most frequently considered quality attributes, and recurring challenges in the formalization, acquisition, and long-term maintenance of domain knowledge. The results provide support for researchers and practitioners in identifying established approaches and unresolved issues. The study also outlines promising directions for future research, emphasizing the development of scalable, automated, and sustainable solutions to integrate domain knowledge into RE processes. [Conclusion] The study contributes by providing a comprehensive overview that helps to build a conceptual and methodological foundation for knowledge-driven requirements engineering.

cs.SE↗

arXiv2025

Agile Management for Machine Learning: A Systematic Mapping Study

[Context] Machine learning (ML)-enabled systems are present in our society, driving significant digital transformations. The dynamic nature of ML development, characterized by experimental cycles and rapid changes in data, poses challenges to traditional project management. Agile methods, with their flexibility and incremental delivery, seem well-suited to address this dynamism. However, it is unclear how to effectively apply these methods in the context of ML-enabled systems, where challenges require tailored approaches. [Goal] Our goal is to outline the state of the art in agile management for ML-enabled systems. [Method] We conducted a systematic mapping study using a hybrid search strategy that combines database searches with backward and forward snowballing iterations. [Results] Our study identified 27 papers published between 2008 and 2024. From these, we identified eight frameworks and categorized recommendations and practices into eight key themes, such as Iteration Flexibility, Innovative ML-specific Artifacts, and the Minimal Viable Model. The main challenge identified across studies was accurate effort estimation for ML-related tasks. [Conclusion] This study contributes by mapping the state of the art and identifying open gaps in the field. While relevant work exists, more robust empirical evaluation is still needed to validate these contributions.

cs.SE↗

arXiv2025

Towards Lean Research Inception: Assessing Practical Relevance of Formulated Research Problems

[Context] The lack of practical relevance in many Software Engineering (SE) research contributions is often rooted in oversimplified views of industrial practice, weak industry connections, and poorly defined research problems. Clear criteria for evaluating SE research problems can help align their value, feasibility, and applicability with industrial needs. [Goal] In this paper, we introduce the Lean Research Inception (LRI) framework, designed to support the formulation and assessment of practically relevant research problems in SE. We describe its initial evaluation strategy conducted in a workshop with a network of SE researchers experienced in industry-academia collaboration and report the evaluation of its three assessment criteria (valuable, feasible, and applicable) regarding their importance in assessing practical relevance. [Method] We applied LRI retroactively to a published research paper, engaging workshop participants in discussing and assessing the research problem by applying the proposed criteria using a semantic differential scale. Participants provided feedback on the criteria's importance and completeness, drawn from their own experiences in industry-academia collaboration. [Results] The findings reveal an overall agreement on the importance of the three criteria - valuable (83.3%), feasible (76.2%), and applicable (73.8%) - for aligning research problems with industrial needs. Qualitative feedback suggested adjustments in terminology with a clearer distinction between feasible and applicable, and refinements for valuable by more clearly considering business value, ROI, and originality. [Conclusion] While LRI constitutes ongoing research and requires further evaluation, our results strengthen our confidence that the three criteria applied using the semantic differential scale can already help the community assess the practical relevance of SE research problems.

cs.SE↗

arXiv2025

Get on the Train or be Left on the Station: Using LLMs for Software Engineering Research

The adoption of Large Language Models (LLMs) is not only transforming software engineering (SE) practice but is also poised to fundamentally disrupt how research is conducted in the field. While perspectives on this transformation range from viewing LLMs as mere productivity tools to considering them revolutionary forces, we argue that the SE research community must proactively engage with and shape the integration of LLMs into research practices, emphasizing human agency in this transformation. As LLMs rapidly become integral to SE research - both as tools that support investigations and as subjects of study - a human-centric perspective is essential. Ensuring human oversight and interpretability is necessary for upholding scientific rigor, fostering ethical responsibility, and driving advancements in the field. Drawing from discussions at the 2nd Copenhagen Symposium on Human-Centered AI in SE, this position paper employs McLuhan's Tetrad of Media Laws to analyze the impact of LLMs on SE research. Through this theoretical lens, we examine how LLMs enhance research capabilities through accelerated ideation and automated processes, make some traditional research practices obsolete, retrieve valuable aspects of historical research approaches, and risk reversal effects when taken to extremes. Our analysis reveals opportunities for innovation and potential pitfalls that require careful consideration. We conclude with a call to action for the SE research community to proactively harness the benefits of LLMs while developing frameworks and guidelines to mitigate their risks, to ensure continued rigor and impact of research in an AI-augmented future.

cs.SE↗

arXiv2025

Who "Controls" Where Work Shall be Done? State-of-Practice in Post-Pandemic Remote Work Regulation

The COVID-19 pandemic has permanently altered workplace structures, making remote work a widespread practice. While many employees advocate for flexibility, many employers reconsider their attitude toward remote work and opt for structured return-to-office mandates. Media headlines repeatedly emphasize that the corporate world is returning to full-time office work. This study examines how companies employing software engineers and supporting roles regulate work location, whether corporate policies have evolved in the last five years, and, if so, how, and why. We collected data on remote work regulation from corporate HR and/or management representatives from 68 corporate entities that vary in size, location, and orientation towards remote or office work. Our findings reveal that although many companies prioritize office-centred working (50%), most companies in our sample permit hybrid working to varying degrees (85%). Remote work regulation does not reveal any particular new "best practice" as policies differ greatly, but the single most popular arrangement was the three in-office days per week. More than half of the companies (51%) encourage or mandate office days, and more than quarter (28%) have changed regulations, gradually increasing the mandatory office presence or implementing differentiated conditions. Although no companies have increased flexibility, only four companies are returning to full-time office work. Our key recommendation for office-oriented companies is to consider a trust-based alternative to strict office presence mandates, while for companies oriented toward remote working, we warn about the points of no (or hard) return. Finally, the current state of policies is clearly not final, as companies continue to experiment and adjust their work regulation.

cs.SE↗

arXiv2025

Designing a Syllabus for a Course on Empirical Software Engineering

Increasingly, courses on Empirical Software Engineering research methods are being offered in higher education institutes across the world, mostly at the M.Sc. and Ph.D. levels. While the need for such courses is evident and in line with modern software engineering curricula, educators designing and implementing such courses have so far been reinventing the wheel; every course is designed from scratch with little to no reuse of ideas or content across the community. Due to the nature of the topic, it is rather difficult to get it right the first time when defining the learning objectives, selecting the material, compiling a reader, and, more importantly, designing relevant and appropriate practical work. This leads to substantial effort (through numerous iterations) and poses risks to the course quality. This chapter attempts to support educators in the first and most crucial step in their course design: creating the syllabus. It does so by consolidating the collective experience of the authors as well as of members of the Empirical Software Engineering community; the latter was mined through two working sessions and an online survey. Specifically, it offers a list of the fundamental building blocks for a syllabus, namely course aims, course topics, and practical assignments. The course topics are also linked to the subsequent chapters of this book, so that readers can dig deeper into those chapters and get support on teaching specific research methods or cross-cutting topics. Finally, we guide educators on how to take these building blocks as a starting point and consider a number of relevant aspects to design a syllabus to meet the needs of their own program, students, and curriculum.

cs.SE↗

arXiv2025

Large Language Model for Qualitative Research -- A Systematic Mapping Study

The exponential growth of text-based data in domains such as healthcare, education, and social sciences has outpaced the capacity of traditional qualitative analysis methods, which are time-intensive and prone to subjectivity. Large Language Models (LLMs), powered by advanced generative AI, have emerged as transformative tools capable of automating and enhancing qualitative analysis. This study systematically maps the literature on the use of LLMs for qualitative research, exploring their application contexts, configurations, methodologies, and evaluation metrics. Findings reveal that LLMs are utilized across diverse fields, demonstrating the potential to automate processes traditionally requiring extensive human input. However, challenges such as reliance on prompt engineering, occasional inaccuracies, and contextual limitations remain significant barriers. This research highlights opportunities for integrating LLMs with human expertise, improving model robustness, and refining evaluation methodologies. By synthesizing trends and identifying research gaps, this study aims to guide future innovations in the application of LLMs for qualitative analysis.

cs.CL↗

arXiv2025

Investigating Issues that Lead to Code Technical Debt in Machine Learning Systems

[Context] Technical debt (TD) in machine learning (ML) systems, much like its counterpart in software engineering (SE), holds the potential to lead to future rework, posing risks to productivity, quality, and team morale. Despite growing attention to TD in SE, the understanding of ML-specific code-related TD remains underexplored. [Objective] This paper aims to identify and discuss the relevance of code-related issues that lead to TD in ML code throughout the ML workflow. [Method] The study first compiled a list of 34 potential issues contributing to TD in ML code by examining the phases of the ML workflow, their typical associated activities, and problem types. This list was refined through two focus group sessions involving nine experienced ML professionals, where each issue was assessed based on its occurrence contributing to TD in ML code and its relevance. [Results] The list of issues contributing to TD in the source code of ML systems was refined from 34 to 30, with 24 of these issues considered highly relevant. The data pre-processing phase was the most critical, with 14 issues considered highly relevant. Shortcuts in code related to typical pre-processing tasks (e.g., handling missing values, outliers, inconsistencies, scaling, rebalancing, and feature selection) often result in "patch fixes" rather than sustainable solutions, leading to the accumulation of TD and increasing maintenance costs. Relevant issues were also found in the data collection, model creation and training, and model evaluation phases. [Conclusion] We have made the final list of issues available to the community and believe it will help raise awareness about issues that need to be addressed throughout the ML workflow to reduce TD and improve the maintainability of ML code.

cs.SE↗

arXiv2025

Can Machine Learning Support the Selection of Studies for Systematic Literature Review Updates?

[Background] Systematic literature reviews (SLRs) are essential for synthesizing evidence in Software Engineering (SE), but keeping them up-to-date requires substantial effort. Study selection, one of the most labor-intensive steps, involves reviewing numerous studies and requires multiple reviewers to minimize bias and avoid loss of evidence. [Objective] This study aims to evaluate if Machine Learning (ML) text classification models can support reviewers in the study selection for SLR updates. [Method] We reproduce the study selection of an SLR update performed by three SE researchers. We trained two supervised ML models (Random Forest and Support Vector Machines) with different configurations using data from the original SLR. We calculated the study selection effectiveness of the ML models for the SLR update in terms of precision, recall, and F-measure. We also compared the performance of human-ML pairs with human-only pairs when selecting studies. [Results] The ML models achieved a modest F-score of 0.33, which is insufficient for reliable automation. However, we found that such models can reduce the study selection effort by 33.9% without loss of evidence (keeping a 100% recall). Our analysis also showed that the initial screening by pairs of human reviewers produces results that are much better aligned with the final SLR update result. [Conclusion] Based on our results, we conclude that although ML models can help reduce the effort involved in SLR updates, achieving rigorous and reliable outcomes still requires the expertise of experienced human reviewers for the initial screening phase.

cs.SE↗

arXiv2025

Impostor Phenomenon Among Software Engineers: Investigating Gender Differences and Well-Being

Research shows that more than half of software professionals experience the Impostor Phenomenon (IP), with a notably higher prevalence among women compared to men. IP can lead to mental health consequences, such as depression and burnout, which can significantly impact personal well-being and software professionals' productivity. This study investigates how IP manifests among software professionals across intersections of gender with race/ethnicity, marital status, number of children, age, and professional experience. Additionally, it examines the well-being of software professionals experiencing IP, providing insights into the interplay between these factors. We analyzed data collected through a theory-driven survey (n = 624) that used validated psychometric instruments to measure IP and well-being in software engineering professionals. We explored the prevalence of IP in the intersections of interest. Additionally, we applied bootstrapping to characterize well-being within our field and statistically tested whether professionals of different genders suffering from IP have lower well-being. The results show that IP occurs more frequently in women and that the prevalence is particularly high among black women as well as among single and childless women. Furthermore, regardless of gender, software engineering professionals suffering from IP have significantly lower well-being. Our findings indicate that effective IP mitigation strategies are needed to improve the well-being of software professionals. Mitigating IP would have particularly positive effects on the well-being of women, who are more frequently affected by IP.

cs.SE↗

arXiv2025

Embracing Experiential Learning: Hackathons as an Educational Strategy for Shaping Soft Skills in Software Engineering

In recent years, Software Engineering (SE) scholars and practitioners have emphasized the importance of integrating soft skills into SE education. However, teaching and learning soft skills are complex, as they cannot be acquired passively through raw knowledge acquisition. On the other hand, hackathons have attracted increasing attention due to their experiential, collaborative, and intensive nature, which certain tasks could be similar to real-world software development. This paper aims to discuss the idea of hackathons as an educational strategy for shaping SE students' soft skills in practice. Initially, we overview the existing literature on soft skills and hackathons in SE education. Then, we report preliminary empirical evidence from a seven-day hybrid hackathon involving 40 students. We assess how the hackathon experience promoted innovative and creative thinking, collaboration and teamwork, and knowledge application among participants through a structured questionnaire designed to evaluate students' self-awareness. Lastly, our findings and new directions are analyzed through the lens of Self-Determination Theory, which offers a psychological lens to understand human behavior. This paper contributes to academia by advocating the potential of hackathons in SE education and proposing concrete plans for future research within SDT. For industry, our discussion has implications around developing soft skills in future SE professionals, thereby enhancing their employability and readiness in the software market.

cs.SE↗

arXiv2025

Towards Emotionally Intelligent Software Engineers: Understanding Students' Self-Perceptions After a Cooperative Learning Experience

[Background] Emotional Intelligence (EI) can impact Software Engineering (SE) outcomes through improved team communication, conflict resolution, and stress management. SE workers face increasing pressure to develop both technical and interpersonal skills, as modern software development emphasizes collaborative work and complex team interactions. Despite EI's documented importance in professional practice, SE education continues to prioritize technical knowledge over emotional and social competencies. [Objective] This paper analyzes SE students' self-perceptions of their EI after a two-month cooperative learning project, using Mayer and Salovey's four-ability model to examine how students handle emotions in collaborative development. [Method] We conducted a case study with 29 SE students organized into four squads within a project-based learning course, collecting data through questionnaires and focus groups that included brainwriting and sharing circles, then analyzing the data using descriptive statistics and open coding. [Results] Students demonstrated stronger abilities in managing their own emotions compared to interpreting others' emotional states. Despite limited formal EI training, they developed informal strategies for emotional management, including structured planning and peer support networks, which they connected to improved productivity and conflict resolution. [Conclusion] This study shows how SE students perceive EI in a collaborative learning context and provides evidence-based insights into the important role of emotional competencies in SE education.

cs.SE↗

arXiv2025

Assessing the Use of AutoML for Data-Driven Software Engineering

Background. Due to the widespread adoption of Artificial Intelligence (AI) and Machine Learning (ML) for building software applications, companies are struggling to recruit employees with a deep understanding of such technologies. In this scenario, AutoML is soaring as a promising solution to fill the AI/ML skills gap since it promises to automate the building of end-to-end AI/ML pipelines that would normally be engineered by specialized team members. Aims. Despite the growing interest and high expectations, there is a dearth of information about the extent to which AutoML is currently adopted by teams developing AI/ML-enabled systems and how it is perceived by practitioners and researchers. Method. To fill these gaps, in this paper, we present a mixed-method study comprising a benchmark of 12 end-to-end AutoML tools on two SE datasets and a user survey with follow-up interviews to further our understanding of AutoML adoption and perception. Results. We found that AutoML solutions can generate models that outperform those trained and optimized by researchers to perform classification tasks in the SE domain. Also, our findings show that the currently available AutoML solutions do not live up to their names as they do not equally support automation across the stages of the ML development workflow and for all the team members. Conclusions. We derive insights to inform the SE research community on how AutoML can facilitate their activities and tool builders on how to design the next generation of AutoML technologies.

cs.SE↗