Communication Challenges and Practices in AI-based Software Systems: An Exploratory Study
Resumo
The integration of AI-based systems into software systems poses technical and organizational challenges, especially regarding communication between multidisciplinary teams. This qualitative study, grounded in the Socio-Technical Grounded Theory, investigates communication practices during the development and integration of AI-based systems. Conducted in two cycles with a total of 15 semistructured interviews, the study identifies five key communication challenges: lack of shared technical vocabulary, misaligned expectations and priorities, unstructured communication practices, and inadequate use of tools. These issues often lead to misunderstandings, rework, and miscoordination across teams. To address them, participants recommended strategies such as shared glossaries, regular alignment meetings, and standardized collaborative tools. These findings contribute to the understanding of socio-technical dynamics in ML-based software projects and offer practical recommendations to improve integration workflows and guide future Software Engineering practices.
Palavras-chave:
AI-based system integration, Software Engineering, Team Collaboration, Communication Practices, ML Project Challenges
Referências
C. Bogart, C. Kästner, J. Herbsleb, and F. Thung. 2021. When and how to make breaking changes: Policies and practices in 18 open source software ecosystems. ACM Transactions on Software Engineering and Methodology 30, 4 (2021), 1–56.
Simon Brandstädter and Kh. Sonntag. 2016. Interdisciplinary Collaboration – How to Foster the Dialogue Across Disciplinary Borders? In Advances in Ergonomic Design of Systems, Products and Processes, Barbara Deml, Patricia Stock, Ralf Bruder, and Christopher M. Schlick (Eds.). Springer, 395–409. DOI: 10.1007/978-3-662-48661-0_35
L. Dabbish, C. Stuart, J. Tsay, and J. Herbsleb. 2012. Social Coding in GitHub: Transparency and Collaboration in an Open Software Repository. In Proc. Conf. Computer Supported Cooperative Work (CSCW). 1277–1286.
C.R.B. de Souza and D.F. Redmiles. 2008. An Empirical Study of Software Developers’ Management of Dependencies and Changes. In Proc. Int’l Conf. Software Engineering (ICSE). 241–250.
Rashina Hoda. 2022. Socio-Technical Grounded Theory for Software Engineering. IEEE Transactions on Software Engineering 48, 10 (2022), 3808–3832. DOI: 10.1109/TSE.2021.3106280
F. Ishikawa and N. Yoshioka. 2019. How Do Engineers Perceive Difficulties in Engineering of Machine-Learning Systems? - Questionnaire Survey. In 2019 IEEE/ACM Joint 7th International Workshop on Conducting Empirical Studies in Industry (CESI) e 6th International Workshop on Software Engineering Pesquisa de Engenharia e Prática Industrial (SERIP). IEEE, 2–9. DOI: 10.1109/CESSER-IP.2019.00009
Christian Kästner. 2025. Machine Learning in Production: From Models to Products. MIT Press, Cambridge, MA. [link] All author royalties donated to Evidence Action. Released under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License. Corresponds to CMU course 17-645 / 11-695..
I. Mistrík, J. Grundy, A. van der Hoek, and J. Whitehead. 2010. Collaborative Software Engineering. Springer.
Nadia Nahar, Shurui Zhou, Grace Lewis, and Christian Kästner. 2022. Collaboration challenges in building ml-enabled systems: Communication, documentation, engineering, and process. In Proceedings of the 44th international conference on software engineering. 413–425.
Ipek Ozkaya. 2020. What Is Really Different in Engineering AI-Enabled Systems? IEEE Software 37, 4 (July/August 2020), 3–6. DOI: 10.1109/MS.2020.2993662
Sheila Maria Mendes Paiva. 2025. Artifacts of the Study on Communication Practices in AI-Based Software Engineering Projects (v1.0). Dataset. DOI: 10.5281/zenodo.15860313
D. Piorkowski et al. 2021. How AI Developers Overcome Communication Challenges in a Multidisciplinary Team: A Case Study. In Proc. ACM on Human-Computer Interaction, Vol. 5. 1–25.
Alex Serban, Koen van der Blom, Holger Hoos, and Joost Visser. 2020. Adoption and Effects of Software Engineering Best Practices in Machine Learning. In Proceedings of the 14th ACM / IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM) (Bari, Italy) (ESEM ’20). Association for Computing Machinery, New York, NY, USA, Article 3, 12 pages. DOI: 10.1145/3382494.3410681
J. Tranquillo. 2017. The T-Shaped Engineer. Journal of Engineering Education Transformations 30, 4 (2017), 12–24.
Simon Brandstädter and Kh. Sonntag. 2016. Interdisciplinary Collaboration – How to Foster the Dialogue Across Disciplinary Borders? In Advances in Ergonomic Design of Systems, Products and Processes, Barbara Deml, Patricia Stock, Ralf Bruder, and Christopher M. Schlick (Eds.). Springer, 395–409. DOI: 10.1007/978-3-662-48661-0_35
L. Dabbish, C. Stuart, J. Tsay, and J. Herbsleb. 2012. Social Coding in GitHub: Transparency and Collaboration in an Open Software Repository. In Proc. Conf. Computer Supported Cooperative Work (CSCW). 1277–1286.
C.R.B. de Souza and D.F. Redmiles. 2008. An Empirical Study of Software Developers’ Management of Dependencies and Changes. In Proc. Int’l Conf. Software Engineering (ICSE). 241–250.
Rashina Hoda. 2022. Socio-Technical Grounded Theory for Software Engineering. IEEE Transactions on Software Engineering 48, 10 (2022), 3808–3832. DOI: 10.1109/TSE.2021.3106280
F. Ishikawa and N. Yoshioka. 2019. How Do Engineers Perceive Difficulties in Engineering of Machine-Learning Systems? - Questionnaire Survey. In 2019 IEEE/ACM Joint 7th International Workshop on Conducting Empirical Studies in Industry (CESI) e 6th International Workshop on Software Engineering Pesquisa de Engenharia e Prática Industrial (SERIP). IEEE, 2–9. DOI: 10.1109/CESSER-IP.2019.00009
Christian Kästner. 2025. Machine Learning in Production: From Models to Products. MIT Press, Cambridge, MA. [link] All author royalties donated to Evidence Action. Released under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License. Corresponds to CMU course 17-645 / 11-695..
I. Mistrík, J. Grundy, A. van der Hoek, and J. Whitehead. 2010. Collaborative Software Engineering. Springer.
Nadia Nahar, Shurui Zhou, Grace Lewis, and Christian Kästner. 2022. Collaboration challenges in building ml-enabled systems: Communication, documentation, engineering, and process. In Proceedings of the 44th international conference on software engineering. 413–425.
Ipek Ozkaya. 2020. What Is Really Different in Engineering AI-Enabled Systems? IEEE Software 37, 4 (July/August 2020), 3–6. DOI: 10.1109/MS.2020.2993662
Sheila Maria Mendes Paiva. 2025. Artifacts of the Study on Communication Practices in AI-Based Software Engineering Projects (v1.0). Dataset. DOI: 10.5281/zenodo.15860313
D. Piorkowski et al. 2021. How AI Developers Overcome Communication Challenges in a Multidisciplinary Team: A Case Study. In Proc. ACM on Human-Computer Interaction, Vol. 5. 1–25.
Alex Serban, Koen van der Blom, Holger Hoos, and Joost Visser. 2020. Adoption and Effects of Software Engineering Best Practices in Machine Learning. In Proceedings of the 14th ACM / IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM) (Bari, Italy) (ESEM ’20). Association for Computing Machinery, New York, NY, USA, Article 3, 12 pages. DOI: 10.1145/3382494.3410681
J. Tranquillo. 2017. The T-Shaped Engineer. Journal of Engineering Education Transformations 30, 4 (2017), 12–24.
Publicado
22/09/2025
Como Citar
PAIVA, Sheila Maria Mendes; ALVES, Arthur; ARAÚJO, Narallynne; MASSONI, Tiago; RAMALHO, Franklin.
Communication Challenges and Practices in AI-based Software Systems: An Exploratory Study. In: SIMPÓSIO BRASILEIRO DE ENGENHARIA DE SOFTWARE (SBES), 39. , 2025, Recife/PE.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 790-796.
ISSN 2833-0633.
DOI: https://doi.org/10.5753/sbes.2025.11574.
