The next phase of architectural modernization: Agents

In 2025, significant advancements in application modernization were driven by cloud innovation and Generative AI (GenAI), which accelerated development processes but also highlighted challenges such as technical debt in complex systems. The introduction of tools like the vFunction Agent marked a shift towards architecture-driven modernization, enabling AI to act based on architectural intent while providing developers with oversight and control. Enhancements included architecture-aware prompts for code assistants, real-time documentation, and a refined approach to managing dependencies and risk, particularly in large, mission-critical applications. With the release of vFunction 4.5, agents now integrate with platforms like Amazon Bedrock and support both relational and non-relational databases, such as MongoDB, to offer a comprehensive view of data usage and dependencies. This evolution laid the groundwork for an increasingly autonomous modernization process in 2026, where AI-driven agents perform tasks within architecturally defined boundaries, enhancing speed and reliability while maintaining user confidence and control.