AWS: Why agentic AI marks an inflection point for enterprise modernization
[The content of this article has been produced by our advertising partner.]
In my decade at AWS, working with hundreds of enterprise customers on their most critical transformations, I haven’t seen anything quite like what we’re experiencing today. Customers understand that using AI can give them competitive advantages in time-to-market, business intelligence, and innovation velocity, but they also recognize that moving their mission-critical applications and data to the cloud is the essential first step in unlocking these advantages.
Let me share what I’m seeing in the field and how this new approach is already transforming how our customers think about modernization—not as a one-time event, but as a continuous competitive advantage.
Just last month, I was working with a large enterprise customer facing a crisis that’s becoming all too common. They had 6,000 virtual machines running critical business applications, and their VMware licensing was about to expire. Beyond simply moving servers, they needed to migrate complex networking configurations and security policies that had been built up over years. Anyone who’s been hands-on with these migrations knows the drill: You spend weeks putting together spreadsheets of IP address blocks, security permissions, and network dependencies. This tedious, error-prone work can derail entire projects.
Three layers that accelerate your transformation timeline
That’s exactly why we built AWS Transform differently. When we launched on May 15, 2025, we didn’t just create another automation tool. We built the world’s first agentic AI service for enterprise transformation, and the difference between this and traditional automation changes everything about what’s possible.
Most people think of AI in enterprise transformation as either simple automation or generic chatbots. But AWS Transform operates on three interconnected layers that work together:
Layer 1: The first layer is our specialized AI agents. These aren’t generic scripts; they’re digital specialists trained for specific transformation challenges. They can modernize decades-old mainframe programming languages (like COBOL) into modern languages (like Java), migrate complex data center network configurations from VMware environments to AWS cloud infrastructure, and update .NET applications to current framework versions.
Layer 2: Above that foundation sits our agentic AI intelligent coordination system that manages the workflow, and this is where the transformation accelerates. This intelligent coordinator understands your business goals and can autonomously manage complex workflows, making decisions about which expert agents to deploy, when to run processes in parallel, and how to handle dependencies that would normally require human intervention. When my customer with 6,000 VMs needed their network configurations migrated, the agentic AI simultaneously orchestrated network discovery agents, security mapping agents, and validation agents. What traditionally takes two weeks of manual spreadsheet work was completed in eight hours with higher accuracy than any human team could achieve.
Layer 3: The third layer brings everything together through natural language chat interfaces that let technical and business teams work together. Migration projects are never solo efforts, and our natural language interfaces translate technical complexity into clear business terms, helping to break down the traditional silos between business stakeholders, architects, and developers. Teams can work together in real-time, making complex technical decisions accessible to everyone involved in the transformation.
My vision for AWS Transform is to make modernization continuous rather than a one-time event. New versions of frameworks keep coming. Security requirements evolve. Business needs change. And our new composability capability allows partners to integrate their own industry-specific data, purpose-built agents, and internal knowledge, creating an environment where adaptation becomes automatic rather than disruptive.
The results we’re seeing across different modernization scenarios demonstrate just how transformative this technology has become. 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.
Air Canada’s experience with custom modernization is equally compelling. Facing high technical debt across thousands of Lambda functions using end-of-life runtimes, their platform team deployed AWS Transform to upgrade from Node.js 16 to 20 runtime in just a few days, achieving a 90% efficacy rate and an 80% reduction in expected time and costs. This represents a fundamental shift in how organizations can break free from decades of legacy infrastructure that has constrained their innovation and growth.
Even as agentic AI handles increasingly complex technical challenges, the human element becomes more critical, not less. The most successful transformations happen when we combine AI’s execution capabilities with human expertise to understand customer problems, synthesize patterns across similar challenges, and develop business strategies that drive successful outcomes. This partnership allows teams to focus on strategic decisions while AI handles the execution.
The technology is ready. The proven results are there from customers like Experian and CSL. The economic case is clear from the dramatic improvements we’re seeing in both speed and cost.
AWS Transform is fundamentally different from anything else in the market—the first agentic AI service that helps reduce technical debt holding back innovation and makes your entire technology stack ready for AI. By clearing away legacy systems and maintenance overhead, you can finally redirect resources toward building the AI capabilities that will define your competitive future.
The window for competitive advantage is open now, but it won’t stay open forever.