At the re:Invent 2025 event, AWS announced a significant enhancement to its application modernization service, "AWS Transform," which introduces a new Agentic AI feature to address the long-standing "Tech Debt" problem faced by enterprises, namely the manual modernization work that enterprise IT teams often need to spend an additional 30% of their time on.
Due to changes in technology upgrades, infrastructure, and hardware environment, enterprise services often require IT professionals to spend a lot of time manually migrating data, switching service platforms, and even rewriting existing service code to ensure that services can operate smoothly in a more modern environment. However, this often becomes a nightmare for many IT professionals.
This updated AWS Transform significantly simplifies the time and resources previously required by IT professionals through agent-based AI technology. AWS states that the new features of AWS Transform not only accelerate the modernization of Windows .NET, VMware, and mainframes, but also enhance "custom" capabilities, allowing enterprises to modernize specific programming languages, frameworks, or even proprietary syntax, thereby saving more time and resources and enabling them to focus more on innovation.
Customizable transformation capabilities: 5 times faster, supports any coding style
The biggest highlight of this update is AWS Transform's custom capability. In the past, enterprises may have many programming languages and frameworks written for specific needs or even outdated ones, which are often difficult to convert using standard tools. Now, through dedicated agent-based AI capabilities, the system can learn and execute consistent and high-quality code conversion, supporting the conversion and upgrade of common programming languages, including Java, Node.js, and Python. Even enterprise-specific programming content can be successfully converted.
Official data shows that for most organizations, AWS Transform Custom can speed up the modernization of hundreds or even thousands of applications up to five times faster than traditional manual operations, while the agent AI will continue to optimize itself through feedback mechanisms.
Actual case:
• Air Canada:Using this tool, the modernization of thousands of Lambda function codes was coordinated and executed in just a few days, reducing project execution time and costs by 80%.
• QAD, an enterprise resource planning software company:The software company stated that a modernization process that previously took two weeks can now be completed in just three days, saving more than 7500 development hours annually.
• Thomson Reuters:By leveraging agent-based AI automation, we have achieved a migration rate of 150 million run-code entries per month, while reducing costs by 30%.
Windows environment fully modernized: shifting to open source, saving 70% in costs.
For enterprises reliant on the Microsoft ecosystem, AWS Transform currently offers full-stack Windows environment modernization acceleration capabilities. Agent-based AI first analyzes the complete Windows software stack (including .NET applications, SQL Server databases, UI frameworks, and operating systems) and proposes a coordinated and consistent transformation plan.
Once approved, Agent AI will perform the transformation, migrating the system to open-source and cloud-native alternatives (such as migrating from a Microsoft SQL Server environment to an open-architecture PostgreSQL). AWS emphasizes that this feature not only improves the efficiency of modernization but also helps enterprise customers save up to 70% on operating costs by eliminating expensive software licensing agreements.
Mainframe and VMware Migration Enhancement
For traditional mainframe architectures, the new AWS Transform features add three AI agent services, including those that can deeply analyze code, extract business rules, and automatically generate technical documentation. In addition, the new task agent can automatically generate test plans and automation scripts, solving the testing problem that previously consumed at least half of IT staff's project time.
Regarding VMware migration, the new agent-based AI feature simplifies large-scale asset inventory and network migration planning processes, supports complete programming layout from assessment to deployment, and can handle unstructured data input (such as files and chat logs) to plan migration work. This feature also enhances support for network architectures and security configurations such as Cisco ACI, Fortigate, and Palo Alto Networks.
In addition, AWS also launched the AWS Transform Composability program, which allows partners such as Accenture, Capgemini, and Pegasystems to integrate their proprietary tools and knowledge bases into AWS Transform, thereby helping specific industries such as finance to provide more contextualized transformation workflows.
