AWS used this year’s re:Invent to take aim at one of the biggest, least glamorous problems in enterprise tech; legacy code. The company introduced AWS Transform, a sweeping effort to automate code modernisation using AI and if it works as advertised, it could change how decades-old systems finally migrate to the cloud.
Legacy modernization is the kind of challenge most executives want to ignore, but almost every major institution faces. Banks still run COBOL systems from the 1970s. Insurance firms operate monolithic Java stacks that predate the iPhone. Governments rely on brittle systems that no one on the current payroll even knows how to maintain. Millions of lines of “mystery code” keep the world running, and rewriting them manually can cost tens or hundreds of millions of dollars.
Transform is essentially Amazon’s attempt to build an AI-powered factory for re-platforming old code. Feed it your Java, your C++, your .NET, your unstructured documentation and AWS claims it can analyse, rewrite, test, and modernize these systems for the cloud. Instead of multi-year migration projects, AWS is pitching a future where machines do most of the heavy lifting while human engineers supervise.
Early demos show AI models automatically untangling interdependent modules, breaking monoliths into services, generating new tests, and even recommending cloud-native architectures. It’s the sort of thing systems integrators and consultants typically charge billions for, now potentially turned into a subscription service operating at cloud scale.
The implications for enterprises are obvious. If even half of this works, companies could shed decades of tech debt faster, adopt cloud-native architectures without astronomical cost, and direct human talent toward innovation rather than maintenance.
But Transform also signals something more strategic for AWS. With Azure pushing aggressively into enterprise AI through Copilot and Microsoft’s longstanding enterprise reach, AWS needed a big counterpunch. An AI-assisted modernization platform fits perfectly into AWS’s strengths: massive cloud infrastructure, deep developer tools, and a huge install base of aging enterprise workloads.
If AWS can prove Transform works in real deployments not just controlled demos, it may become one of the most consequential announcements of the entire AI era.
Paul Balo is the founder of TechBooky and a highly skilled wireless communications professional with a strong background in cloud computing, offering extensive experience in designing, implementing, and managing wireless communication systems.
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.
