AWS Transform – AI-Powered Code Modernization

What is AWS Transform?

• AWS Transform is an agentic AI-powered service that modernizes enterprise workloads at scale — including full-stack Windows/.NET applications, mainframe systems, VMware infrastructure, and custom code transformations.
• It evolved from AWS’s migration and modernization tools (including AWS Migration Hub, AWS Schema Conversion Tool, and Porting Assistant for .NET) into a unified, AI-driven platform.
• The service uses specialized task agents built on decades of migration experience combined with enterprise-specific context.
• Agents use goal-driven orchestration ranging from deterministic execution to dynamic plans, with humans in the loop for oversight.
• Learning capability is built-in at every level — agents continually self-debug, improve outcomes, and provide recommendations.
• Available through a unified web experience, CLI, IDE integrations (Visual Studio, Kiro, Claude Code, Cursor), and MCP server.
• Supports collaborative workspaces where architects define target states, developers execute, leads review, and partners deliver at scale.

AWS Transform Key Capabilities

1. AWS Transform for .NET

• Purpose: Modernize .NET Framework applications to cross-platform .NET (e.g., .NET 8) that runs on Linux.
• First agentic AI service for modernizing .NET applications at scale — launched GA in May 2025.
• Ports entire applications including dependencies — handles MVC, WCF, Web APIs, and console applications.
• Automates code analysis, dependency mapping, compatibility assessment, and refactoring tasks.
• Accelerates .NET modernization by up to 4x compared to traditional manual approaches.
• Reduces Windows licensing costs by up to 40% by enabling Linux deployment.
• Applications run 1.5–2x faster with improved performance and 50% better scalability on Linux.
• Includes a conversational AI assistant for Visual Studio for developer-level application work.
• Supports deployment to Amazon EC2 Linux, Amazon ECS, Amazon EKS, and AWS Lambda.
• Customer Example: Experian modernized 7 legacy .NET applications (687,600 lines of code), saving ~300 engineering days with ~40% developer effort reduction.
• Customer Example: Signaturit Group cut Windows .NET to Linux migration from 6-8 months to a few days.

2. AWS Transform for Mainframe

• Purpose: Modernize mainframe workloads (COBOL, PL/I, JCL) to cloud-native applications.
• Supports multiple modernization patterns: Refactor (automated code conversion) and Reimagine (business logic extraction → cloud-native redesign).
• Reimagine Capabilities:
  • Extracts business rules from legacy COBOL/PL/I code with full traceability
  • Converts to syntax-independent specifications
  • Generates cloud-native Java microservices with REST APIs and entity mappings
  • Every requirement traces back to source code for auditable transformation decisions
• Automated Testing: Generates test cases, test data collection scripts, and test automation scripts for validation.
• Supports IBM z/OS COBOL, VSAM, IMS, DB2, and expanded to PL/I (common in financial services and insurance).
• Connected assessment-to-code-generation workflow compresses months of discovery into hours.
• Native integration with Kiro IDE — developers steer forward engineering conversationally.
• Automates analysis of mainframe codebases: JCL, BMS, COBOL programs, and copybooks.
• Customer Example: BMW Group reduced test case creation from 10 days to hours, increased test coverage by 60%, and migrated 7 applications in 6 months — targeting 12-month reduction in overall transformation timeline.

3. AWS Transform for SQL Server

• Purpose: Modernize SQL Server databases to Amazon Aurora PostgreSQL — the successor to AWS Schema Conversion Tool (SCT).
• Accelerates SQL Server to Aurora PostgreSQL modernization by up to 5x through intelligent schema conversion.
• Handles the complete migration lifecycle:
  • Schema analysis and conversion
  • Stored procedure transformation to PostgreSQL-compatible format
  • Application code refactoring (Entity Framework configs, connection strings)
  • Data migration
• Three layers of validation: syntax validation, semantic equivalence, and functional verification with synthetic data.
• Supports virtual sources so teams don’t need direct production database access to start.
• Iterative workflow: get an assessment with level of effort → DBAs review and approve → Transform executes.
• Coordinates database modernization with application code changes simultaneously.

4. AWS Transform Custom

• Purpose: Learn your organization’s specific patterns and automate transformations across repositories at scale.
• Transforms any code pattern — version upgrades, runtime migrations, framework transitions, language translations, and architecture decompositions.
• Pre-built transformations include: Java upgrades, Node.js upgrades, Python upgrades, boto2→boto3, AWS SDK migrations, x86→Graviton, Spring Boot updates, Angular→React, Vue.js upgrades, Log4j→SLF4J, Progress 4GL→Java, ColdFusion→React/Java, and more.
• Continual Learning: The agent automatically captures patterns, fixes, and edge cases as reusable knowledge items, so transformations get faster and more reliable with every run.
• Define once, transform everywhere — capture transformation knowledge and execute repeatable tasks across your entire organization.
• Up to 85% efficacy rate for out-of-the-box transformations (Java, Node.js upgrades).
• Available via CLI, web experience, Kiro Power, Claude Code, VS Code, and can be embedded in any pipeline.
• Customer Example: Air Canada achieved 90% efficacy rate and 80% reduction in expected time and costs upgrading thousands of Lambda functions from Node.js 16 to 20.
• Customer Example: Twitch achieved 70% acceleration on AWS SDK v1→v2 Golang migration across 913 repositories, saving ~2,876 developer days (11 developer years).
• Customer Example: Coupang transformed 70+ Java applications in 2 months with a team of 5 — a 90% timeline reduction.

5. AWS Transform – Continuous Modernization (Preview, June 2026)

• Purpose: Always-on, autonomous portfolio management that continuously finds tech debt, fixes it, validates, and learns.
• Announced at AWS Summit New York 2026 — shifts code transformation from periodic projects into an automated, pipeline-driven practice (CI/CD/CM — Continuous Modernization).
• Continuous Analysis:
  • Automatically scans code repositories against configurable baselines
  • Generates findings in hours, not weeks
  • Detects end-of-life dependencies, deprecated frameworks, security vulnerabilities
  • Extend with organization-specific policies (approved libraries, internal coding standards)
  • Provides ground truth directly from code — no manual compliance tracking
• Autonomous Remediation at Scale:
  • Generates pull requests for affected repositories automatically
  • Notifies owning teams with context and proposed fix
  • Teams review, merge, or remediate using their own approach
  • Detects when fixes are in place without manual confirmation
• Integrations: GitHub organizations, GitLab groups, Bitbucket workspaces, local repositories, AWS CodePipeline, Jenkins, GitHub Actions.
• Integrates with AWS Security Agent for source-code-level security vulnerability remediation.
• Available through the AWS Transform web application, Kiro Power, or MCP for integration with existing coding agents.

6. AWS Transform for Full-Stack Windows Modernization

• Purpose: Coordinated transformation across all layers — application code, UI framework, database, and deployment.
• Accelerates full-stack Windows modernization by up to 5x using specialized domain-expert agents.
• Reduces operating costs by up to 70% by moving away from costly Windows/SQL Server licenses.
• Four Transformation Layers:
  • Application Layer: .NET Framework → cross-platform .NET (Linux-ready)
  • UI Layer: ASP.NET Web Forms → Blazor (modern, cross-platform)
  • Database Layer: SQL Server → Amazon Aurora PostgreSQL (schema + stored procedures + app code)
  • Deployment Layer: Automated CI/CD pipeline generation, CloudFormation templates, ECS/EC2 Linux deployment
• Unified web experience with natural language interaction for coordinated modernization plans.
• Agents assess complexity, sequence work into waves, and execute transformations end-to-end with human oversight.
• Architects can step in at any point to steer decisions without breaking the autonomous flow.
• Up to 40% better price-performance running modernized apps on AWS Graviton vs. x86 instances.

AWS Transform vs. Manual Refactoring vs. Third-Party Tools

Criteria	Manual Refactoring	Third-Party Tools (Snyk, SonarQube)	AWS Transform
Scope	Single app at a time	Detection + limited auto-fix	Full-stack transformation at scale (code + DB + UI + deployment)
Approach	Developer-driven, line by line	Rule-based scanning + suggestions	Agentic AI with goal-driven orchestration
Speed	Months to years per application	Fast detection, manual remediation	Up to 5x faster end-to-end transformation
Scale	Limited by team size	Portfolio scanning, per-repo fixes	Hundreds of applications in parallel
Learning	Tribal knowledge, inconsistent	Static rule updates	Continual learning from every execution (knowledge items)
Mainframe Support	Specialist consulting required	Not supported	Full COBOL/PL/I → cloud-native with traceability
Database Migration	Manual schema + stored proc conversion	Not supported	Intelligent schema conversion + coordinated app code changes
Continuous Tech Debt	Periodic sprints, reactive	Continuous detection, manual fix	Autonomous detection + remediation + PR generation
Validation	Manual testing	Linting and SAST	Multi-layer: syntax, semantic equivalence, functional verification
Cost Model	Engineering headcount	Per-developer licensing	Pay per transformation job

Customer Results

• Overall Impact: 4.5+ billion lines of code processed, 1.6+ million hours saved (929 developer years), hundreds of thousands of servers migrated in the first year.
• BMW Group: Used AWS Transform for mainframe modernization — reduced test case creation from 10 days to hours, increased test coverage by 60%, migrated 7 applications in 6 months.
• Experian: Modernized 7 .NET Framework applications (687,600 LOC) to .NET 8 using AWS Transform for .NET — saved ~300 engineering days with ~40% developer effort reduction.
• Air Canada: Upgraded thousands of Lambda functions from Node.js 16 to 20 — achieved 90% efficacy rate and 80% reduction in time/costs. Made AWS Transform their internal standard.
• Twitch: AWS SDK v1→v2 Golang migration across 913 repositories — 70% acceleration, saving ~2,876 developer days (11 developer years).
• Coupang: Transformed 70+ Java applications in 2 months with 5 developers — 90% timeline reduction vs. traditional manual approaches.
• CSL: Planned migrations for thousands of servers in days — a 10x acceleration over prior approaches.
• ADP: Modernized complex mainframe using Transform’s mainframe and custom capabilities — now scaling for 1.1 million clients with results in weeks.
• 4 out of 5 customers return to do additional projects; roughly half use multiple transformation capabilities.

How AWS Transform Works – Architecture

• Expert Task Agents: Dozens of specialized agents for network generation, business rule extraction, .NET porting, schema conversion, etc.
• Agentic Orchestration: Goal-driven orchestration that adapts per workload — deterministic where precision is needed, dynamic where flexibility is required.
• Built-in Learning: Knowledge items captured from debugging steps, human input, and code observations improve future executions.
• Human-in-the-Loop: Teams supervise, approve plans, override decisions, and step in/out of autonomous workflows.
• Shared Context: Seamless handoffs between stages — no re-entry, no lost progress across web, CLI, and IDE surfaces.
• Composability: Customers, partners, and ISVs can build custom agents using Agent Builder Toolkit and integrate with AWS Transform via MCP server.

AWS Transform Pricing

• AWS Transform pricing is based on the specific capability used and transformation scope.
• Custom transformations are priced per transformation job.
• Continuous modernization pricing is based on repository connections and remediation volume.
• Some capabilities (like model-to-model migration assessment) are available at no additional charge beyond standard pricing.
• Refer to the AWS Transform Pricing page for current details.

AWS Certification Exam Practice Questions

1. A company wants to modernize 200 .NET Framework applications running on Windows Server to reduce licensing costs and improve performance. Which AWS service should they use to accelerate this transformation?

1. AWS Migration Hub
2. AWS App2Container
3. AWS Transform for .NET
4. AWS Elastic Beanstalk

Show Answer

Answer: C –

Explanation: AWS Transform for .NET is specifically designed to modernize .NET Framework applications to cross-platform .NET at scale, accelerating modernization by up to 4x and reducing Windows licensing costs by up to 40%.

2. An enterprise is modernizing a legacy COBOL mainframe system. They need to convert business logic into cloud-native microservices while maintaining full traceability from source to target. Which AWS Transform capability should they use?

1. AWS Transform Custom
2. AWS Transform for .NET
3. AWS Transform for Mainframe – Reimagine
4. AWS Transform – Continuous Modernization

Show Answer

Answer: C –

Explanation: AWS Transform for Mainframe’s Reimagine capability extracts business rules from COBOL/PL/I code with full traceability and generates cloud-native Java microservices with REST APIs, maintaining an audit trail from source to modernized code.

3. A platform engineering team manages 2,000+ repositories and wants to continuously detect and remediate tech debt (end-of-life dependencies, deprecated frameworks) without periodic maintenance sprints. Which capability best fits this requirement?

1. AWS Transform Custom with CLI automation
2. AWS Transform – Continuous Modernization
3. Amazon CodeGuru Reviewer
4. AWS Config Rules

Show Answer

Answer: B –

Explanation: AWS Transform – Continuous Modernization (Preview, June 2026) provides always-on, autonomous tech debt analysis and remediation at scale. It continuously scans repositories, generates prioritized findings, and autonomously creates pull requests for remediation — shifting from periodic projects to CI/CD/CM.

4. A company needs to upgrade Java versions, migrate AWS SDK v1 to v2, and convert Angular to React across hundreds of applications consistently. They want the transformation agent to learn from each execution and improve over time. Which capability should they use?

1. AWS Transform for Full-Stack Windows Modernization
2. Amazon Q Developer
3. AWS Transform Custom
4. AWS Transform for Mainframe

Show Answer

Answer: C –

Explanation: AWS Transform Custom provides pre-built and custom transformations for diverse code patterns (Java upgrades, SDK migrations, framework transitions). It features continual learning through knowledge items — capturing patterns, fixes, and edge cases from every execution to improve future transformations.

5. An organization is modernizing its Windows technology stack and needs coordinated transformation across .NET applications, ASP.NET Web Forms UI, SQL Server databases, and deployment processes. Which approach provides unified modernization across all layers?

1. Use separate tools: AWS Transform for .NET + AWS DMS + manual UI rewrite
2. AWS Transform for Full-Stack Windows Modernization
3. AWS Elastic Beanstalk with Docker migration
4. AWS Transform Custom with multiple transformation definitions

Show Answer

Answer: B –

Explanation: AWS Transform for Full-Stack Windows Modernization provides coordinated transformation across all four layers — application (.NET → cross-platform), UI (Web Forms → Blazor), database (SQL Server → Aurora PostgreSQL), and deployment (CI/CD pipeline generation). It uses domain-expert agents in a unified experience for cohesive modernization.

Frequently Asked Questions

What is AWS Transform?

AWS Transform is an agentic AI service for large-scale code modernization. It handles .NET Framework to cross-platform .NET, mainframe COBOL to cloud-native, SQL Server migrations, and custom transformations — having eliminated 1.6M+ hours of manual effort for customers like BMW and Experian.

What is Transform Continuous Modernization?

Launched in June 2026, Continuous Modernization is an always-on capability that autonomously monitors your code repositories, identifies tech debt as it accumulates, fixes it, validates the fix, and integrates with your existing CI/CD pipelines (GitHub Actions, Jenkins, GitLab, CodePipeline).

Can AWS Transform modernize mainframe applications?

Yes. Transform for Mainframe can convert COBOL, PL/I, and other legacy code to cloud-native Java or .NET using its Reimagine capability. It also provides automated testing to validate functional equivalence, reducing modernization timelines from years to months.

References

• AWS Transform – Official Page
• AWS Transform for .NET
• AWS Transform for Mainframe
• AWS Transform for SQL Server Modernization
• AWS Transform Custom
• AWS Transform – Continuous Modernization
• AWS Transform for Full-Stack Windows Modernization
• One Year Milestone: 4.5 Billion Lines of Code, 1.6 Million Hours Saved
• Continuous Modernization Preview Announcement