Abstract

The rapid adoption of serverless computing has significantly influenced the design of modern workflow automation systems by offering scalability, operational simplicity, and cost-efficient execution. Nevertheless, the majority of available workflow orchestration solutions are also static and rule-based, which makes them less adaptable to dynamism and uncertainty as well as dynamically to changing business conditions. This paper tries to solve these shortcomings by suggesting a smart serverless workflow automation architecture that combines Agentic AI and cloud-native execution. The designed solution has an independent decision-making layer that can plan and coordinate dynamics of workflows at runtime instead of using preset execution paths. To achieve scalability and resilience workflow tasks are implemented as stateless services, which are deployed on Google Cloud platform using managed serverless services, and implemented as RESTful services written in Java. The research method involves architectural design, practical implementation and use case-based assessment of the feasibility and performance. The results of the evaluation point to the fact that the suggested architecture allows enhancing the flexibility of the workflow and fault tolerance without compromising the scalability and cost efficiency of serverless platforms. Decoupling decision-making and execution, and using Agentic AI to achieve adaptive orchestration, this work provides a generalisable architectural framework of intelligent workflow automation in a cloud setting, and has practical applications in areas where an enterprise system needs to establish dynamic and autonomous process management.

References

• Poorvadevi, R., Surendar, H. and SriRamakrishnan, S., 2025, April. An Operational Exploration of Data Processing on Cloud Platform Using Serverless Computing. In 2025 International Conference on Computing and Communication Technologies (ICCCT) (pp. 1-5). IEEE.
• Aslani, A. and Ghobaei-Arani, M., 2025. Machine learning inference serving models in serverless computing: a survey. Computing, 107(1), p.47.
• Angelis, A. and Kousiouris, G., 2025. An Overview on the Landscape of Self-Adaptive Cloud Design and Operation Patterns: Goals, Strategies, Tooling, Evaluation, and Dataset Perspectives. Future Internet, 17(10), p.434.
• Saleh, S.M., Madhavji, N.H. and Steinbacher, J., 2025, November. Systematic Review of Identity-Centric Security in Cloud-Native CI/CD Pipelines. In Proceedings of the 2025 10th International Conference on Cloud Computing and Internet of Things (pp. 23-32).
• Soussi, W., Gür, G. and Stiller, B., 2024. Democratizing container live migration for enhanced future networks-a survey. ACM Computing Surveys, 57(4), pp.1-37.
• Mongkoljaturong, K., Manitsakulwong, M. and Moodleah, S., 2025. AI-Powered MetaHuman Interviewer: Serious Game for Student Job Interview Skills. IEEE Access, 13, pp.205505-205520.
• Bastidas Fuertes, A., Pérez, M. and Meza Hormaza, J., 2023. Transpilers: A systematic mapping review of their usage in research and industry. Applied Sciences, 13(6), p.3667.
• Xu, C., Du, X., Fan, X., Giuliani, G., Hu, Z., Wang, W., Liu, J., Wang, T., Yan, Z., Zhu, J. and Jiang, T., 2022. Cloud-based storage and computing for remote sensing big data: a technical review. International Journal of Digital Earth, 15(1), pp.1417-1445.
• Alzoubi, Y.I., Al-Ahmad, A., Kahtan, H. and Jaradat, A., 2022. Internet of things and blockchain integration: security, privacy, technical, and design challenges. Future Internet, 14(7), p.216.
• Nastic, S., 2024. Self-provisioning infrastructures for the next generation serverless computing. SN Computer Science, 5(6), p.678.
• Maciá-Lillo, A., Mora, H., Jimeno-Morenilla, A., García-D’Urso, N.E. and Azorín-López, J., 2025. AI edge cloud service provisioning for knowledge management smart applications. Scientific Reports, 15(1), p.32246.
• Liu, Y., Fu, L. and Penghui, M., 2025. ServerlessPGO: enhancing serverless cold starts through PGO. Cluster Computing, 28(14), p.877.
• Golec, M., Ponugoti, S.S. and Gill, S.S., 2024. Enhancing data security for cloud service providers using AI. In Applications of AI for Interdisciplinary Research (pp. 187-204). CRC Press.
• Gonzalez, L.A., Neyem, A., Contreras‐McKay, I. and Molina, D., 2022. Improving learning experiences in software engineering capstone courses using artificial intelligence virtual assistants. Computer Applications in Engineering Education, 30(5), pp.1370-1389.
• Atobatele, O.K., Ajayi, O.O., Hungbo, A.Q. and Adeyemi, C., 2023. Enhancing the accuracy and integrity of immunization registry data using scalable cloud-based validation frameworks. International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 9(5), pp.787-806.
• Shih, C.C., Chen, J., Lee, A.S., Bertin, N., Hebrard, M., Khor, C.C., Li, Z., Juan Tan, J.H., Meah, W.Y., Peh, S.Q. and Mok, S.Q., 2022. RAPTOR: A Five-Safes approach to a secure, cloud native and serverless genomics data repository. bioRxiv, pp.2022-10.
• Wang, L., Yan, J., Ma, Y., Huang, X., Li, J., Wang, S., He, H., Long, A. and Zhang, X., 2024. Cloud computing in remote sensing: A comprehensive assessment of state of the arts. In Remote Sensing Handbook, Volume I (pp. 399-438). CRC Press.
• Kataria, D., Walid, A., Daneshmand, M., Dutta, A., Enright, M.A., Gu, R., Lackpour, A., Ramachandran, P., Wang, H., Chen, C.M. and Chng, B., 2023, November. INGR Roadmap Artificial Intelligence And Machine Learning Chapter. In 2023 IEEE Future Networks World Forum (FNWF) (pp. 1-69). IEEE.
• Bianculli, D., Sartaj, H., Andrikopoulos, V., Pautasso, C., Mikkonen, T., Perez, J., Bureš, T., De Sanctis, M., Muccini, H., Navarro, E. and Soliman, M. eds., 2025. Software Architecture. ECSA 2025 Tracks and Workshops: Limassol, Cyprus, September 15–19, 2025, Proceedings. Springer Nature.
• Nestorov, A.M., Marrón, D., Gutierrez-Torre, A., Wang, C., Misale, C., Youssef, A., Carrera, D. and Berral, J.L., 2024, December. Dexter: a performance-cost efficient resource allocation manager for serverless data analytics. In Proceedings of the 25th International Middleware Conference (pp. 117-130).