Smash tech debt with AWS Transform: The new era of migration and modernization

Technical debt isn’t just a developer inconvenience—it’s a material drag on enterprise innovation. The numbers tell a stark story: Forrester reports that nearly 20% of IT budgets are spent managing technical debt instead of advancing new capabilities. The Journal of Systems and Software found that developers lose an average of 23% of their working time dealing with technical debt. According to Accenture, in the US alone, tech debt costs $2.41 trillion a year and would require $1.52 trillion to fix.

The math is clear: reducing technical debt is one of the fastest ways to accelerate innovation and delivery velocity. Every dollar and hour reclaimed moves directly into new value creation. But reducing tech debt at scale has never been easy—until now.

Eliminate tech debt with agentic AI

AWS Transform is the first agentic AI service built to accelerate enterprise transformation at scale. Built on nearly two decades of AWS experience, it applies AI to increase efficiency and speed, reduce costs, and modernize applications with greater accuracy. AWS Transform accelerates full-stack Windows modernization, mainframe modernization, and VMware migration, as well as custom transformations of code, APIs, frameworks, and more—making tech stacks AI-ready while eliminating the technical debt that slows innovation.

Since we launched the general availability of AWS Transform on May 15, 2025, the momentum has been remarkable. Customers have already used AWS Transform to save 1,009,000 hours of manual effort (the equivalent of 483 developer years of work) while analyzing 1.8 billion lines of code as they migrate and modernize applications in the cloud.

Experian provides a powerful example of this acceleration. Using AWS Transform, the Experian Data Office modernized seven legacy .NET applications, achieving a 40% reduction in developer effort and saving approximately 300 engineering days while transforming hundreds of thousands of lines of code.

Accelerate full-stack Windows modernization

The reality of Windows-based workloads modernization extends far beyond application code. Organizations face interconnected challenges across their entire technology stack—from legacy .NET Framework applications and outdated UI frameworks to SQL Server databases and deployment processes that were designed for a different era. Addressing these layers in isolation creates friction, inconsistency, and ultimately, incomplete transformations that fail to deliver the promised value.

That’s why we didn’t stop at the .NET application layer. We’ve extended support to include SQL Server database modernization to Amazon Aurora PostgreSQL, adding UI Frameworks modernization, as well as helping customers with deployment processes for transformed applications, enabling full-stack Windows modernization.

AWS Transform for full-stack Windows modernization accelerates transformations by up to 5x, transforming complete technology stacks including .NET Framework applications and legacy UI frameworks, SQL Server databases, and deployment processes into cloud-native solutions. With specialized AI transformation agents, modernization teams can collaboratively execute larger and more complex multi-layer projects with consistency, reduce operating costs by up to 70% by moving away from costly licenses, and enhance code quality, performance, and security across all technology layers.

Customers are seeing significant results with Windows modernization. IDEMIA accelerated their application modernization 4x faster, completing in weeks what previously took months, while reducing TCO by 30% and improving their security posture with .NET 8 support. Grupo Tress Internacional (GTI) achieved a 70% reduction in modernization effort across 135,000 lines of code, saving 2 months of team time while reducing infrastructure costs by over 40% using Graviton-based workloads.

Transform any code pattern and crush tech debt

We don’t think about it, but repetition is at the core of modernization. And for many of us it’s a primary reason why we accumulate technical debt in the first place. It’s boring work that no one wants to do. And that boring, repetitive work applies to every application you create—patching, updating versions, upgrading runtimes and more. One of the core reasons customers are achieving breakthrough results is that our specialized AI agent is automating repetitive tasks to help customers tackle technical debt faster than ever before. As customers experienced the power of AWS Transform for things like upgrading .NET frameworks and translating COBOL to Java, they started to ask if Transform could help with the rest of their tech debt backlog. That’s exactly why we built AWS Transform custom.

AWS Transform custom uses agentic AI to accelerate organization-wide code and application modernization at scale through automated transformations, both built-in and custom. It supports diverse scenarios from version upgrades and runtime migrations to complex language translations and architectural changes, reducing execution time by over 80% in many cases. Through continual learning from code samples, documentation, and developer feedback, it delivers high-quality, repeatable transformation tasks while eliminating the need for specialized automation expertise, enabling organizations to scale their modernization initiatives effectively.

This first-of-its-kind, any-to-any agent tackles modernization for code, APIs, frameworks, language runtimes, architectural changes, and language translations to address technical debt. The agent offers out-of-the-box transformations such as Java, Node.js, Python upgrades, while allowing customers to custom-define their own transformations for their specific organizations’ needs. This eliminates specialized automation expertise requirements while democratizing transformation knowledge across organizations, enabling consistent modernization across hundreds of applications simultaneously with continual learning capabilities that improve accuracy over time.

Air Canada’s experience demonstrates the transformative impact. Ray Galipeau, Senior Director, Cloud, Network & Platform Services at Air Canada, shares: “We were struggling with high technical debt for a lot of our codebases, including thousands of Lambda functions using end-of-life runtimes. We needed to upgrade from Node.js 16 to 20 runtime among other modernization efforts. In a few days, our platform team was able to deploy AWS Transform to coordinate and execute the modernization of all of them, achieving a 90% efficacy rate and an 80% reduction in expected time and costs for the project.”

Cut mainframe modernization from years to months

Mainframe modernization has long been the third rail of enterprise IT—necessary but daunting, with projects stretching across years and consuming massive resources. The complexity isn’t just technical; it’s institutional knowledge locked in decades-old code and business logic that’s never been fully documented, and testing requirements that traditionally consume over half the project timeline. The industry needed a fundamentally different approach.

We’re revolutionizing mainframe modernization, cutting project timelines from years to months with agentic AI automating analysis, business logic extraction, documentation generation, decomposition, refactoring, reimagining, and testing. AWS Transform for mainframe now provides code, data and activity analysis capabilities that extract insights to drive application decomposition and inform coding agents like Kiro in reimagining mainframe applications into cloud-native architectures.

Additionally, AWS Transform for mainframe now offers automation capabilities designed to reduce the time and effort required for mainframe modernization testing, which typically consumes over 50% of project duration. This includes automated test plan generation, test data collection scripts, and test case automation scripts, alongside functional test environment tools for continuous delivery and regression testing, helping accelerate and de-risk testing and validation.

BMW Group reduced testing time by 75% and increased test coverage by 60% using AWS Transform for mainframe, significantly lowering risk while accelerating modernization timelines. Itaú cut discovery time by 96%, reduced testing creation effort by 97%, and achieved a 4x improvement in test coverage while accelerating migration speed by 75%.

Simplify VMware migrations at scale

VMware migrations have traditionally been exercises in manual coordination—discovery tools generate data, planning happens separately, execution requires careful sequencing, and network reconfiguration demands deep expertise. Each phase can introduce delays, potential errors, and misalignment between what was planned and what gets deployed.

AWS Transform for VMware introduced new agentic AI capabilities to fundamentally change how to migrate your VMware workloads to AWS. The VMware migration agent becomes your collaborative teammate, understanding your business priorities and intelligently planning migrations for hundreds of applications spanning thousands of servers.

The agent discovers your environment using multiple sources—the AWS Transform discovery tool, discovery data from select third-party tools including Cloudamize, Matilda Cloud, and modelizeIT, and even unstructured information like documents and business rules. It analyzes infrastructure and dependencies, then generates migration plans organized around business and technical priorities: ownership, department, function, subnet, operating systems. It generates hub-and-spoke or isolated network configurations, handles flexible IP address management, deploys across multiple accounts, and migrates from NSX, Palo Alto, Fortigate, and Cisco ACI source environments. Throughout your migration, you can ask the agent questions to guide your decisions, modify plans as needed, and repeat or skip discovery steps to accommodate infrastructure changes. For internal approvals, it generates detailed reports mapping networks, servers, and applications.

Using AWS Transform for VMware, CSL accelerated initial wave planning by 10x, saving a minimum of 10.5 weeks of effort for 1,072 applications across 29 datacenters, while reducing application discovery time by 12x. This resulted in 30% operational cost savings through licensing cost avoidance, hardware costs, and data center related costs.

Leverage industry-specific expertise through partner integrations

Certain complicated transformations often require industry-specific knowledge—regulatory requirements in financial services, compliance patterns in healthcare, or specialized frameworks unique to particular sectors. While AWS Transform provides powerful agentic AI capabilities, we recognized that our partners bring decades of domain expertise that could make transformations even more effective and contextually relevant.

We designed the AWS Transform composability initiative to allow for partner integrations to create what we call composable transformations. This allows our partners to create customized transformation workflows with their own industry-specific data, purpose-built agents, and industry knowledge bases. We’re working with partners like Accenture, Pegasystems, Capgemini, and more to leverage their industry experience, enhance code conversion capabilities, support additional languages, and integrate their purpose-built agents.

Take action on technical debt today

The cost of waiting is too high. With technical debt consuming up to 30% of development resources and costing trillions annually, organizations can no longer afford to let legacy systems hold them back. AWS Transform provides the agentic AI capabilities needed to modernize at scale, whether you’re transforming Windows-based applications, migrating from VMware, modernizing mainframe workloads, or tackling custom code modernization challenges across your entire portfolio.

The results speak for themselves: customers are saving millions of hours and reducing transformation time by over 80% in many cases. They’re reclaiming the resources previously lost to technical debt and redirecting them toward innovation and new value creation.

Learn more about AWS Transform and start your modernization journey today.

For a deeper perspective on how agentic AI marks an inflection point for enterprise modernization, read the article by Dr. Asa Kalavade, VP of Migration & Modernization at AWS.