The Challenge
We Modernised an Application built on .NET Framework 4.8, ran exclusively on Windows Server—driving up infrastructure costs through expensive licensing fees and limiting their ability to optimise cloud infrastructure. The application's monolithic architecture created operational bottlenecks, while the dependency on Windows Server prevented the company from leveraging cost-effective Linux-based deployment options.
The Solution
We Implemented AWS Transform for .NET, the industry's first agentic AI service purpose-built to modernise legacy .NET applications at scale. The architecture leverages generative AI-powered agents with Large Language Models to automate code transformation from .NET Framework to cross-platform .NET, enabling deployment on Linux containers and eliminating Windows Server licensing costs. It achieved significant TCO reduction, transformed 7,870 lines of code (30% of the codebase) within hours, and slashed transformation time from months to weeks.
Architecture Overview
The solution architecture comprises four layers: Core AI and Transformation Engine, Deployment Infrastructure, Security & Compliance, and Monitoring & Operations.
Core AI and Transformation Engine
AWS Transform Agent for .NET provides generative AI-powered refactoring with automated code analysis, transformation supervision, and validation:
- Analysis Phase: Discovered the codebase from source control, identified cross-repository dependencies, analyzed private NuGet packages, and generated comprehensive assessment reports
- Transformation Phase: Automated conversion of the Vendor Management application from .NET Framework 4.8 to .NET 8 through AI-supervised refactoring, updated namespaces and dependency injection configurations, transformed ASP.NET MVC components, and handled service and data layer modernization
- Validation Phase: Executed unit tests on transformed code, validated build success with zero errors, generated transformation summary reports, and committed transformed code to new branches for review
Supported Project Types: The transformation focused on class libraries, console applications, ASP.NET MVC, Web API, and service/data layers, with UI modernization planned as a separate phase to maintain project momentum.
Deployment Infrastructure and Cost Optimization
Amazon ECS - Containerized applications with managed orchestration:
- Linux Container Deployment: Eliminates Windows Server licensing costs entirely
- High Availability Architecture: Spans two Availability Zones for resilience
- Auto Scaling Groups: Adjusts capacity based on demand automatically
- Simplified Operations: Managed container orchestration reduces operational overhead
Cost Savings Architecture
Eliminate Windows Licensing - Primary cost reduction:
- Migrate from .NET Framework on Windows to cross-platform .NET on Linux
- Remove Windows Server licensing fees entirely
- Eliminate version upgrade and maintenance costs
- Achieve significant TCO reduction through infrastructure optimization
Container Optimization - Additional operational efficiency:
- Right-size CPU/memory allocations to avoid over-provisioning
- Implement demand-based auto-scaling to reduce idle resources
- Leverage managed services to reduce operational overhead
- Improve scalability and performance with modern container architecture
Workflow Orchestration
Analysis Stage:
- Analyzed the application codebase and dependencies
- Identified all four core projects requiring transformation
- Mapped connections between service and data layers
- Generated compatibility assessment for .NET 8 migration
Transformation Stage:
- Converted 7,870 lines of code from .NET Framework 4.8 to .NET 8 within hours
- Updated dependencies, namespaces, and framework references
- Transformed service and data layer components
- Handled dependency injection configuration updates
Validation Stage:
- Validated transformed code through comprehensive unit testing
- Ensured successful builds with zero compilation errors
- Generated detailed transformation reports with actionable insights
- Provided developers with valuable insights throughout the process
Deployment Stage:
- Prepared deployment to Amazon ECS using Linux containers
- Configured high availability across two Availability Zones
- Implemented Auto Scaling groups for demand-based capacity adjustment
- Established blueprint for broader modernisation efforts
Security and Compliance
Encryption Standards:
- Data at Rest: AES-256 encryption with customer-managed KMS keys
- Data in Transit: TLS 1.3 for all API communications and repository connections
- Key Rotation: Automated KMS key rotation policies
Access Control:
- IAM Roles: Least-privilege permissions for bedrock:InvokeModel and bedrock:InvokeAgent actions
- Repository Access: Secure CodeConnections with IP allowlisting
- VPC Security: VPC endpoints for private connectivity, security group-based access control
Audit and Compliance:
- AWS CloudTrail: Complete audit trail of all transformation activities and API calls
- Transformation Logs: Detailed logs of AI decisions, code changes, and validation results
- Compliance Validation: Support for SOC 2, ISO 27001, GDPR, CCPA requirements
Monitoring and Operations
Real-time Monitoring:
- Visual Studio Integration: AWS Transformation Hub window with real-time progress, estimated completion times, detailed step-by-step worklogs
Transformation Reports:
- Assessment Reports: Detailed analysis of .NET Core compatibility, NuGet package classifications
- Transformation Detail Reports: Comprehensive change documentation, reasons for modifications, actionable error details
- Developer Insights: Valuable suggestions throughout the transformation process
Measurable Impact
Transformation Speed:
- 7,870 lines of code Transformed within hours (30% of codebase)
- Months to weeks: Slashed transformation time dramatically
- All 4 core projects: Successfully modernized with AI assistance
- Initial adoption: Immediate productivity gains from first use
Cost Savings:
- Significant TCO reduction: Eliminated Windows Server licensing costs
- Infrastructure optimization: Linux containers reduce operational expenses
- Reduced maintenance: Modern .NET 8 framework requires less upkeep
- Scalability improvements: Auto Scaling reduces over-provisioning costs
Operational Benefits:
- Improved performance: Modern .NET 8 delivers better application performance
- Better user experience: Enhanced scalability and responsiveness
- High availability: Two Availability Zone architecture ensures resilience
- Developer productivity: AI-powered insights accelerate development
Strategic Value
- Blueprint established: Repeatable process for broader modernization efforts
- Multiple business units: Framework for enterprise-wide transformation
- Business continuity: Maintained operations throughout modernization
- Cloud optimization: Leveraged cloud-based services for improved efficiency
Conclusion
The .NET Framework modernisation journey demonstrates how AWS Transform's agentic AI capabilities can eliminate the barriers to enterprise application modernisation while delivering substantial cost savings and operational improvements. By automating code transformation from .NET Framework 4.8 to cross-platform .NET 8, We transformed 7,870 lines of code within hours, slashed transformation time from months to weeks, and achieved significant TCO reduction through eliminated Windows Server licensing.
The architecture's AI-supervised approach minimizes manual effort while providing developers with valuable insights throughout the transformation process, enabling organizations to focus on innovation rather than undifferentiated heavy lifting. As organizations face mounting pressure from licensing costs, technical debt accumulation, and the need for cloud optimization, The Hartford's experience provides a proven pathway to modernization with measurable financial and operational impact. The combination of AWS Transform's intelligent automation, Amazon ECS's container orchestration, and AWS's scalable infrastructure delivers a comprehensive solution for .NET modernization at enterprise scale.