AWS Transform Agentic AI for Cloud Migration
AWS Transform: Agentic AI is redefining enterprise infrastructure migration
At Re:Invent 2025, AWS unveiled major innovations in AWS Transform, introducing agentic AI capabilities that fundamentally change how enterprises migrate and modernize critical infrastructure. This technological leap addresses a persistent challenge: 75% of enterprise workloads remain on-premises, with 70% of Fortune 500 companies still running software that is more than 20 years old.
For Québec and Canadian organizations, this announcement comes at a critical moment. Rising VMware licensing costs, the shortage of mainframe talent due to retirements, and the imperative to move to the cloud to leverage AI are creating unprecedented pressure. AWS Transform promises to accelerate transitions that traditionally required 18 months into a matter of weeks or months, while significantly reducing risk and manual effort.
The promise of agentic AI: Automating human expertise
AWS Transform represents a fundamental shift in cloud migration. Where traditional approaches require months of manual analysis, dependency mapping, and detailed planning, AWS Transform deploys specialized AI agents that collaborate directly with migration teams.
These agents do more than automate repetitive tasks: they understand business context, analyze complex dependencies, and make informed decisions throughout the migration process. A conversational agent allows teams to ask questions, adjust plans on the fly, and repeat or skip steps as needed.
The results speak for themselves: AWS Transform has already analyzed 1.1 billion lines of code and saved more than 810,000 hours of manual work for customers. According to announcements at Re:Invent 2025, the service can now accelerate transformations up to 5 times faster than manual approaches.
Three modernization pillars: VMware, mainframe, and .NET
VMware migration: breaking the cost spiral
The AWS Transform agent for VMware directly addresses rising VMware licensing costs. The service automates the entire migration pipeline:
Automated discovery and inventory
- New on-premises discovery tool that scans the existing VMware environment
- Support for integrating inventories from third-party tools
- Analysis of unstructured data (documents, notes, business rules)
Priority-based intelligent planning
- The agent understands business and technical priorities
- Automatic generation of migration plans grouped by department, function, subnet, or operating system
- Use of Graph Neural Networks to analyze network traffic and identify dependencies
- Creation of optimized wave-based migration plans
Advanced network migration
- Hub-and-spoke configurations and isolated networks
- Support for advanced security technologies from Cisco ACI, Fortigate, and Palo Alto Networks
- Flexible IP address management
- Deployment across multiple AWS accounts
- Secure, iterative migration with continuous progress updates
Broad compatibility
- Windows and Linux x86
- Multiple hypervisors: VMware, HyperV, Nutanix, KVM
- Physical bare-metal environments
- Deployment across 16 AWS Regions
Mainframe modernization: Beyond lift-and-shift
The mainframe remains the backbone of many financial institutions and large enterprises. AWS Transform for Mainframe introduces breakthrough capabilities:
Application Reimagining
AI agents can now fully reimagine mainframe applications rather than simply migrate them. This approach breaks down monoliths into services aligned with business domains, generating a modern, cloud-native architecture.
Test automation
Test creation, which traditionally took days, is now automated in hours. BMW reported a 75% time and efficiency gain, along with a 60% increase in test coverage.
Business rule reverse engineering
The agent automatically analyzes and documents undocumented business rules buried in decades of COBOL or PL/I code, a task that was previously close to impossible given the shortage of mainframe expertise.
.NET transformation: Reducing Windows licensing costs
The AWS Transform agent for .NET accelerates porting .NET Framework applications from Windows to Linux, delivering up to 40% operational cost savings. These savings come from:
- Reduced Windows Server licensing costs
- Elimination of complex version upgrades
- Lower maintenance costs
- Avoidance of end-of-support issues
The agent automatically analyzes dependencies, applies domain knowledge from previous modernizations, transforms code autonomously, runs unit tests, and validates Linux compatibility.
Real-world use cases: Measurable results
Air Canada: 80% reduction in time and cost
Air Canada, the country’s largest airline, needed to modernize outdated software threatening its 24/7 operations. In just days, Air Canada deployed AWS Transform to coordinate and execute the modernization of thousands of Lambda functions, achieving an 80% reduction in time and cost compared to a manual migration.
BMW: Transforming the migration factory
BMW used the agentic capabilities of AWS Transform in its “migration factory” (an expert team augmented by AI). The results:
- Test creation time reduced from days to hours (75% savings)
- 60% increase in test coverage
- Resources freed to focus on innovation instead of maintenance
Mercedes-Benz: Migrating a global ordering system
Mercedes-Benz used generative AI and AWS agentic refactoring to modernize its global mainframe-based ordering system. The project converted 1.3 million lines of COBOL code into Java, fundamentally transforming the company’s critical infrastructure.
Michael Hermann, Head of Global SAP Strategy at Mercedes-Benz, stated:
“We migrated our largest application, the group’s target financial system OneERP, in just nine months, a record timeline for an organization with tens of thousands of users.”
Danske Bank: Reimagining core banking workloads
Danske Bank uses AWS Transform and Kiro (AWS’s AI IDE) to analyze and reimagine its critical mainframe workloads, with AI-assisted code analysis, reverse engineering of undocumented business rules, and decomposition of monoliths into domain-aligned services.
Technical architecture: How agentic AI works
AWS Transform relies on several sophisticated technical components working together:
Domain-specialized agents
Each agent is trained for a specific transformation type:
- Discovery agent: scans and inventories the existing environment
- Planning agent: analyzes dependencies and generates optimized plans
- Network migration agent: converts complex network configurations
- Refactoring agent: transforms application code
- Test agent: automatically generates and executes tests
Continuous learning
Agents automatically capture feedback and improve over time. Each subsequent transformation becomes more reliable and efficient, creating cumulative expertise.
Human-in-the-loop
Despite deep automation, AWS Transform maintains essential human oversight. Teams can:
- Review and approve generated artifacts
- Adjust plans during execution
- Ask the agent questions to guide decisions
- Repeat or skip steps as needed
This approach ensures human expertise remains central, while AI removes repetitive, time-consuming tasks.
The composability initiative: Partner ecosystem
A major Re:Invent 2025 announcement was the AWS Transform Composability initiative. This platform allows AWS partners to integrate proprietary tools, agents, and knowledge bases to build custom workflows within the AWS Transform product experience.
This open model creates an ecosystem where:
- Specialized partners can contribute domain expertise
- Customers benefit from customized agents tailored to their needs
- Industry knowledge can be embedded directly into agents
- Collaborative innovation accelerates the entire industry
Business impact and ROI
Drastic reduction in time-to-value
Traditional migration projects typically span 18 months or more. AWS Transform compresses this timeline into weeks or months, enabling organizations to:
- Accelerate AI adoption on modern cloud foundations
- Reduce infrastructure costs during transition
- Redirect IT budgets toward innovation rather than maintenance
Freeing technical resources
Organizations typically spend 30% of team time on manual modernization work, the well-known “technical debt.” This work is necessary but diverts valuable resources from value-creating innovation.
By automating this workload, AWS Transform enables teams to:
- Focus on developing new features
- Accelerate AI and machine learning initiatives
- Improve customer experience rather than maintain legacy systems
Risk reduction
Manual migration of critical infrastructure carries significant risk:
- Human errors in dependency mapping
- Overlooked critical configurations
- Misunderstood business rules buried in legacy code
- Late discovery of compatibility issues
AWS Transform’s AI agents systematically analyze every component, automatically identify dependencies, and validate configurations, significantly reducing the risk of disruption.
Considerations for Québec and Canadian SMBs
Accessibility for mid-sized organizations
Although AWS Transform is designed for large-scale enterprise migrations, its automation-based model makes it accessible to growing SMBs:
Predictable costs
Automation reduces reliance on costly external consultants. Organizations can begin with internal teams augmented by AI agents.
Progressive scalability
SMBs can start by migrating non-critical workloads to validate the approach, then gradually extend to critical systems.
Multi-region support
With support for Canada Central (Montréal) and Canada West (Calgary), Québec and Canadian companies can keep data within national borders while leveraging advanced AWS Transform capabilities.
Data sovereignty implications
AWS Transform’s multi-region capabilities make it possible to structure migrations that meet data residency requirements:
- Migration to Canadian Regions (ca-central-1 in Montréal or ca-west-1 in Calgary)
- Full control over data location throughout the migration
- Compliance with Canadian sector regulations
Preparing your organization
To maximize the value of AWS Transform, organizations should:
- Accurately inventory the current environment
- Document critical applications
- Identify key dependencies
- Assess current technical debt
- Define business priorities
- Which applications should migrate first?
- Which systems can be reimagined versus simply migrated?
- What is the available budget and expected ROI?
- Prepare teams
- Train teams on cloud-native principles
- Establish cloud governance processes
- Define roles and responsibilities during migration
The future of migration: Toward full autonomy
The Re:Invent 2025 announcements mark a fundamental shift in cloud migration. The industry is moving from manual, project-based approaches to a continuous, largely automated model.
AWS also introduced three “frontier agents,” a new class of AI agents capable of operating autonomously for hours or days:
- Kiro Autonomous Agent: End-to-end software development
- AWS Security Agent: Proactive security and threat detection
- AWS DevOps Agent: Operations automation
These agents represent the natural evolution of AWS Transform: systems that not only migrate existing infrastructure, but actively participate in its design, deployment, and ongoing optimization.
Conclusion: The time to act
Several converging factors create a unique window of opportunity for Québec and Canadian organizations:
- Rising VMware licensing costs are forcing infrastructure reassessment
- Mainframe talent shortages make modernization urgent
- AI adoption requires modern cloud foundations
- AWS Transform’s agentic AI capabilities make the transition faster and less risky than ever
Organizations that delay modernization accumulate technical debt and lose competitiveness against more agile competitors. Conversely, those that embrace these capabilities can turn migration constraints into innovation opportunities.
With proven results from leaders such as Air Canada, BMW, and Mercedes-Benz, AWS Transform demonstrates that large-scale modernization is not only possible, it can be achieved in a fraction of the traditional time with controlled risk.
For SMBs and enterprises across Québec and Canada, the equation is clear: invest in modernization today with agentic AI tools, or face the growing consequences of technical inertia in the years ahead.
Need support with your cloud migration strategy? At Unicorne, we help Québec organizations navigate the complexity of critical infrastructure migrations. Our expertise in AWS cloud architecture and cost optimization enables us to design migration strategies that maximize ROI while minimizing risk.
Contact us for a personalized infrastructure assessment and discover how the new AWS Transform capabilities could accelerate your digital transformation.
If you want to learn more:
AWS. (2025). “AWS Transform adds new agentic AI capabilities for enterprise VMware migrations”. AWS What’s New. https://aws.amazon.com/about-aws/whats-new/2025/12/transform-vmware-agentic-ai-enterprise-migration/
About Amazon. (2025). “AWS Transform adds AI agents to modernize codes and apps”. AWS re:Invent 2025 News. https://www.aboutamazon.com/news/aws/aws-transform-ai-agents-windows-modern
AWS Migration & Modernization Blog. (2025). “Transform Enterprise Workloads up to 4x Faster with Agentic AI”. https://aws.amazon.com/blogs/migration-and-modernization/aws-transform-generally-available/
AWS. (2025). “AWS for mainframe modernization – re:Invent 2025 Refresher”. AWS Migration & Modernization Blog. https://aws.amazon.com/blogs/migration-and-modernization/aws-for-mainframe-modernization-reinvent-2025-refresher/
AWS Industries Blog. (2025). “Financial institutions advance mission-critical workloads and Agentic AI at re:Invent 2025”. https://aws.amazon.com/blogs/industries/financial-institutions-advance-mission-critical-workloads-and-agentic-ai-at-reinvent-2025/
AWS. (2025). “Mercedes-Benz moving to AWS for RISE with SAP and Agentic AI”. AWS Case Studies. https://aws.amazon.com/solutions/case-studies/mercedes-benz-transform-case-study/
AWS Industries Blog. (2025). “AWS re:Invent 2025 Recap for Automotive and Manufacturing”. https://aws.amazon.com/blogs/industries/aws-reinvent-2025-recap-for-automotive-and-manufacturing/
About Amazon. (2025). “AWS re:Invent 2025: Amazon announces Nova 2, Trainium3”. https://www.aboutamazon.com/news/aws/aws-re-invent-2025-ai-news-updates