Agentes basados en LLMs para la derivación de tareas de desarrollo backend a partir de requerimientos textuales
Resumo
El desarrollo backend presenta desafíos para transformar requisitos funcionales informales en tareas precisas e implementables. Este trabajo explora el uso de Modelos de Lenguaje de Gran Escala (LLMs) como agentes colaborativos y orientados a tareas para apoyar las etapas tempranas del ciclo de vida del software. Se propone un flujo de trabajo modular multi-agente en el que un conjunto de agentes especializados asisten a un desarrollador a refinar requisitos, descomponer funcionalidades en servicios, definir contratos de API, especificar suites de pruebas y asegurar consistencia. El objetivo es reducir el esfuerzo manual, la sobrecarga de coordinación y posibles desalineamientos, mejorando la velocidad y calidad del desarrollo.Referências
Ahmad, A., Waseem, M., Liang, P., Fahmideh, M., Aktar, M. S., and Mikkonen, T. (2023). Towards human-bot collaborative software architecting with chatgpt. arXiv:2302.14600.
Bachmann, F., Bass, L., Klein, M., and Shelton., C. (2005). Designing software architectures to achieve quality attribute requirements. IEE Proceedings - Software.
Bass, L., Clements, P., Kazman, R., and a. O. M. C. Safari (2021). Software Architecture in Practice, 4th Edition. Addison-Wesley Professional.
Capilla, R., Jansen, A., Tang, A., Avgeriou, P., and Babar, M. (2016). 10 years of software architecture knowledge management: Practice and future. Journal of Systems and Software 116.
Dhar, R., Vaidhyanathan, K., and Varma, V. (2024). Can llms generate architectural design decisions?-an exploratory empirical study. arXiv:2403.01709.
Diaz-Pace, J. A., Tommasel, A., Capilla, R., and Ramírez, Y. E. (2025). Architecture exploration and reflection meet llm-based agents. 2025 IEEE 22nd International Conference on Software Architecture Companion (ICSA-C).
Díaz-Pace, J. A., Tommasel, A., and Capilla, R. (2024). Helping novice architects to make quality design decisions using an llm-based assistant. European Conference on Software Architecture.
Ferrari, R. and Madhavji, N. H. (2008). Software architecting without requirements knowledge and experience: What are the repercussions? Journal of Systems and Software.
Jimenez, C. E., Yang, J., Wettig, A., Yao, S., Pei, K., Press, O., and Narasimhan., K. R. (2024). Swe-bench: Can language models resolve real-world github issues? The Twelfth International Conference on Learning Representations.
Ozkaya, I. (2024). Can architecture knowledge guide software development with generative ai? IEEE Software, vol. 40, no. 5.
Sami, M. A., Waseem, M., Zhang, Z., Rasheed, Z., Systä, K., and Abrahamsson., P. (2024). Ai based multiagent approach for requirements elicitation and analysis. arXiv:2409.00038.
Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., Zhao, W. X., Wei, Z., and Wen., J. (2024). A survey on large language model based autonomous agents. Fronts. of Comp. Science.
White, J., Hays, S., Fu, Q., Spencer-Smith, J., and Schmidt., D. C. (2024). ChatGPT Prompt Patterns for Improving Code Quality, Refactoring, Requirements Elicitation, and Software Design. Springer Nature Switzerland.
Yao, S., Zhao, J., Yu, D., Du, N., Shafran, I., Narasimhan, K., and Cao, Y. (2022). React: Synergizing reasoning and acting in language models. arXiv:2210.03629.
Zhou, A., Yan, K., Shlapentokh-Rothman, M., Wang, H., and Wang, Y. X. (2023). Language agent tree search unifies reasoning acting and planning in language models. arXiv:2310.04406.
Bachmann, F., Bass, L., Klein, M., and Shelton., C. (2005). Designing software architectures to achieve quality attribute requirements. IEE Proceedings - Software.
Bass, L., Clements, P., Kazman, R., and a. O. M. C. Safari (2021). Software Architecture in Practice, 4th Edition. Addison-Wesley Professional.
Capilla, R., Jansen, A., Tang, A., Avgeriou, P., and Babar, M. (2016). 10 years of software architecture knowledge management: Practice and future. Journal of Systems and Software 116.
Dhar, R., Vaidhyanathan, K., and Varma, V. (2024). Can llms generate architectural design decisions?-an exploratory empirical study. arXiv:2403.01709.
Diaz-Pace, J. A., Tommasel, A., Capilla, R., and Ramírez, Y. E. (2025). Architecture exploration and reflection meet llm-based agents. 2025 IEEE 22nd International Conference on Software Architecture Companion (ICSA-C).
Díaz-Pace, J. A., Tommasel, A., and Capilla, R. (2024). Helping novice architects to make quality design decisions using an llm-based assistant. European Conference on Software Architecture.
Ferrari, R. and Madhavji, N. H. (2008). Software architecting without requirements knowledge and experience: What are the repercussions? Journal of Systems and Software.
Jimenez, C. E., Yang, J., Wettig, A., Yao, S., Pei, K., Press, O., and Narasimhan., K. R. (2024). Swe-bench: Can language models resolve real-world github issues? The Twelfth International Conference on Learning Representations.
Ozkaya, I. (2024). Can architecture knowledge guide software development with generative ai? IEEE Software, vol. 40, no. 5.
Sami, M. A., Waseem, M., Zhang, Z., Rasheed, Z., Systä, K., and Abrahamsson., P. (2024). Ai based multiagent approach for requirements elicitation and analysis. arXiv:2409.00038.
Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., Zhao, W. X., Wei, Z., and Wen., J. (2024). A survey on large language model based autonomous agents. Fronts. of Comp. Science.
White, J., Hays, S., Fu, Q., Spencer-Smith, J., and Schmidt., D. C. (2024). ChatGPT Prompt Patterns for Improving Code Quality, Refactoring, Requirements Elicitation, and Software Design. Springer Nature Switzerland.
Yao, S., Zhao, J., Yu, D., Du, N., Shafran, I., Narasimhan, K., and Cao, Y. (2022). React: Synergizing reasoning and acting in language models. arXiv:2210.03629.
Zhou, A., Yan, K., Shlapentokh-Rothman, M., Wang, H., and Wang, Y. X. (2023). Language agent tree search unifies reasoning acting and planning in language models. arXiv:2310.04406.
Publicado
11/05/2026
Como Citar
PALERMO, Nicolás Miccio.
Agentes basados en LLMs para la derivación de tareas de desarrollo backend a partir de requerimientos textuales. In: CONGRESSO IBERO-AMERICANO EM ENGENHARIA DE SOFTWARE (CIBSE), 29. , 2026, Recife/PE.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 349-356.
