Case Study
Eliminating Risk from a Critical Legacy Application
A leading automotive manufacturer needed to modernize a business-critical legacy application carrying significant security and technical debt. Infosys piloted an AI-assisted engineering approach using Anthropic Claude Code to remediate vulnerabilities, upgrade the application stack and strengthen test coverage, while keeping engineers responsible for review, validation and sign-off throughout the modernization journey.
Key Challenges
Mounting Security and Compliance Risk
The application carried 281 high and critical vulnerabilities alongside 1,351 quality issues on an unsupported legacy technology stack.
A Complex Modernization Path
Moving from JDK 11 through JDK 17 to JDK 21, alongside Spring modernization and workflow-engine changes, created a high-risk upgrade path.
Limited Test Coverage
With only 9% test coverage, large-scale remediation carried significant regression risk and made conventional modernization harder to execute confidently.
Ready to experience?
Talk To ExpertsThe Solution
Modernizing Legacy Applications with Governed AI Engineering
Strengthening the Engineering Safety Net
- Increased automated test coverage before undertaking major application changes.
- Used mutation-tested validation to create a stronger foundation for controlled modernization.
- Improved the ability to detect regressions as AI-assisted changes were introduced.
AI-assisted Remediation and Modernization
- Used Claude Code to execute application upgrades and security fixes in governed, reviewable batches.
- Modernized the application from JDK 11 to JDK 21 with Spring 6 and removed the legacy workflow engine.
- Applied AI-assisted reasoning to diagnose complex dependency and API issues during modernization.
Human-in-the-Loop Governance
- Kept Infosys engineers responsible for pull-request review, validation and sign-off at every milestone.
- Structured AI-generated changes into logical batches to maintain control and traceability.
- Combined autonomous engineering assistance with human oversight across the full remediation process.
From Legacy Risk to an AI-ready Application Core
- AI-assisted engineering accelerates complex remediation while keeping every major change subject to structured human review.
- Security vulnerabilities and quality issues can be addressed alongside application modernization rather than through separate transformation programs.
- A governed, repeatable approach creates a stronger foundation for extending AI-assisted modernization across other applications carrying similar technical debt.
Business Outcomes
281 to zero
high and critical security vulnerabilities
1,351 to zero
SONAR quality issues
9% to 69%
increase in JUnit test coverage
~30 %
reduction in engineering effort
Creating a Repeatable Path to Legacy Modernization
The pilot demonstrates how governed AI-assisted engineering can accelerate legacy modernization while strengthening security, quality and maintainability. By combining autonomous code analysis and remediation with structured human oversight, the manufacturer modernized a high-risk application without sacrificing control. The approach also provides a repeatable model for addressing similar security and technology debt across a broader application estate.
Request for services
Find out more about how we can help your organization navigate its next. Let us know your areas of interest so that we can serve you better.
