Autonomous AI Agents for Mainframe Modernization: A Framework for COBOL-to-Cloud Banking Systems

Legacy mainframe systems continue to serve as the technological backbone of global financial institutions due to their reliability, scalability, and transaction-processing capabilities. However, the increasing demand for cloud-native architectures, digital banking services, and operational agility has accelerated the need for modernizing COBOL-based banking applications. Traditional modernization approaches are often labor-intensive, costly, and susceptible to errors because of undocumented business rules, tightly coupled dependencies, and decades of accumulated technical debt. This paper proposes an Autonomous AI Agent Framework for COBOL-to-Cloud Banking System Modernization that integrates artificial intelligence-driven software analysis with automated migration assistance. The proposed framework employs specialized AI agents to perform source code understanding, dependency discovery, business rule extraction, data flow analysis, documentation generation, migration planning, and validation of transformed components. The framework further incorporates large language model-assisted code interpretation with static analysis techniques to improve maintainability while preserving business logic integrity. A conceptual evaluation based on representative banking modernization scenarios demonstrates the framework's potential to reduce manual effort, improve migration accuracy, accelerate documentation generation, and minimize modernization risks. The study highlights the practical applicability of autonomous AI agents in enterprise legacy transformation and provides a structured roadmap for financial institutions planning gradual migration from COBOL-based environments to cloud-native architectures.

Mainframe Modernization, COBOL, Cloud Migration, Autonomous AI Agents, Legacy Systems, Banking Systems, Software Modernization, Enterprise

G. Acampora, V. Loia, and A. Vitiello, “Enhancing ontology alignment through a memetic aggregation of similarity measures,” Inf. Sci., vol. 250, pp. 1–20, Nov. 2013.

G. D'Angelo, D. Della-Morte, D. Pastore, G. Donadel, A. De Stefano, and F. Palmieri, “Identifying patterns in multiple biomarkers to diagnose diabetic foot using an explainable genetic programming-based approach,” Future Gener. Comput. Syst., vol. 140, pp. 138–150, Mar. 2023.

N. Ferranti, S. S. R. F. Soares, and J. F. de Souza, “Metaheuristics-based ontology meta-matching approaches,” Expert Syst. Appl., vol. 173, pp. 1–15, Jul. 2021.

M. Freitag, G. Foster, D. Grangier, V. Ratnakar, Q. Tan, and W. Macherey, “Experts, errors, and context: A large-scale study of human evaluation for machine translation,” Trans. Assoc. Comput. Linguist., vol. 9, pp. 1460–1474, Dec. 2021.

J. R. Koza, “Genetic programming as a means for programming computers by natural selection,” Statist. Comput., vol. 4, no. 2, pp. 87–112, 1994.

G. Lan, T. Liu, X. Wang, X. Pan, and Z. Huang, “A semantic Web technology index,” Sci. Rep., vol. 12, no. 1, pp. 1–10, 2022.

H. Li, “User validation in ontology alignment: Functional assessment and impact,” Knowl. Eng. Rev., vol. 34, pp. 1–33, Jan. 2019.

Z. Lv and R. Peng, “A novel periodic learning ontology matching model based on interactive grasshopper optimization algorithm,” Knowl.-Based Syst., vol. 228, pp. 1–14, Sep. 2021.

Y. Ma, S. Zhao, W. Wang, Y. Li, and I. King, “Multimodality in meta-learning: A comprehensive survey,” Knowl.-Based Syst., vol. 250, pp. 1–21, Aug. 2022.

J. Martinez-Gil and J. F. Aldana-Montes, “Evaluation of two heuristic approaches to solve the ontology meta-matching problem,” Knowl. Inf. Syst., vol. 26, no. 2, pp. 225–247, 2011.

J. Martinez-Gil and J. F. Aldana-Montes, “An overview of current ontology meta-matching solutions,” Knowl. Eng. Rev., vol. 27, no. 4, pp. 393–412, 2012.

J. Martinez-Gil and J. Chaves-González, “A novel method based on symbolic regression for interpretable semantic similarity measurement,” Expert Syst. Appl., vol. 160, pp. 1–46, Dec. 2020.

J. Martinez-Gil, E. Alba, and J. F. Aldana-Montes, “Optimizing ontology alignments by using genetic algorithms,” in Proc. Workshop Nature Based Reason. Semant. Web., Karlsruhe, Germany, 2008, pp. 1–14.

J. Martinez-Gil, S. Yin, J. Küng, and F. Morvan, “Matching large biomedical ontologies using symbolic regression,” in Proc. 23rd Int. Conf. Inf. Integr. Web Intell., 2021, pp. 162–167.

E. Mosqueira-Rey, E. Hernández-Pereira, D. Alonso-Ríos, J. Bobes-Bascarán, and Á. Fernández-Leal, “Human-in-the-loop machine learning: A state of the art,” Artif. Intell. Rev., vol. 56, no. 4, pp. 3005–3054, 2023.

S. K. Narayanasamy, K. Srinivasan, Y.-C. Hu, S. K. Masilamani, and K.-Y. Huang, “A contemporary review on utilizing semantic Web technologies in healthcare, virtual communities, and ontology-based information processing systems,” Electronics, vol. 11, no. 3, pp. 1–28, 2022.

J. Portisch, M. Hladik, and H. Paulheim, “Background knowledge in ontology matching: A survey,” Semant. Web, vol. 15, no. 6, pp. 1–55, 2022.

A. Singh and A. Sharan, “Adaptive genetic programming based linkage rule miner for entity linking in semantic Web,” in Proc. Int. Conf. Comput., Commun. Autom. (ICCCA), 2017, pp. 373–378.

J. Sousa, “Zero-defect manufacturing terminology standardization: Definition, improvement, and harmonization,” Front. Manuf. Technol., vol. 2, pp. 1–14, Dec. 2022.

A. Swar, G. Khoriba, and M. Belal, “A unified ontology-based data integration approach for the Internet of Things,” Int. J. Elect. Comput. Eng., vol. 12, no. 2, p. 2097, 2022.

P. Wang, S. Zou, J. Liu, and W. Ke, “Matching biomedical ontologies with GCN-based feature propagation,” Math. Biosci. Eng, vol. 19, no. 8, pp. 8479–8504, 2022.

J. Wang, Z. Ding, and C. Jiang, “GAOM: Genetic algorithm based ontology matching,” in Proc. IEEE Asia–Pac. Conf. Services Comput. (APSCC), 2006, pp. 617–620.

Z. Zheng, B. Zhou, D. Zhou, A. Q. Khan, A. Soylu, and E. Kharlamov, “Towards a statistic ontology for data analysis in smart manufacturing,” in Proc. ISWC Posters, Demos Ind. Tracks, 2022, pp. 1–5