Case Study: Achieving 50% Faster Legacy Modernisation with AI-Driven Engineering

22 Apr 2026 · Last updated: 21 Apr 2026
Natalie Gray, Director of Marketing & Growth

Natalie Gray, Director of Marketing & Growth

See author's bio and posts

Prefer listening over reading? Press play and enjoy

Case Study: Achieving 50% Faster Legacy Modernisation with AI-Driven Engineering
7:15

When a business-critical platform relies on technology that has been unsupported for decades, the risks to security and compliance are no longer theoretical; they are imminent. Many organisations feel trapped between what they expect is a slow, expensive project of a full manual rebuild and the stagnation of technical debt.

This article explores how Codurance navigated this challenge for a large enterprise, moving them off a legacy VB6 environment within a constrained timeframe. We delve into our AI-accelerated delivery model, showing how a 'human-in-the-loop' approach can turn an 18-month projected modernisation project into a delivery measured in just a few months, all while maintaining total business continuity. 

About the Client

Our client is a large enterprise operating in a highly regulated environment. Its operations depend on a business-critical internal platform used across multiple teams to manage core operational workflows.

This platform was deeply embedded in the organisation but built using Visual Basic 6 (VB6), a legacy technology that is no longer supported. This created increasing operational, security and compliance risks, while also limiting the organisation’s ability to evolve the system.

The Challenge

An audit identified that the core application was built on VB6, which had been unsupported for years. This created a clear need to modernise within a defined timeframe.

The client needed to:

  • Move away from unsupported VB6 technology quickly
  • Maintain business continuity across a widely used system
  • Preserve existing functionality and user experience
  • Deliver within a constrained budget and timeline

An initial proposal to rebuild the system as a modern web application using React was estimated to take 18+ months, making it unviable commercially and operationally.

Additionally, knowledge of the system was concentrated in a single internal developer, increasing delivery risk and limiting long-term sustainability.

What we Delivered

1. Migration from VB6 to C# .NET

One of the most effective techniques was introducing a quality gate pipeline.

We delivered a full conversion of the legacy application from VB6 to C# .NET WinForms, preserving existing functionality, workflows and user experience.

The scope was intentionally like-for-like, including maintaining existing behaviours and even visual elements where required, to minimise disruption and ensure continuity.

2. AI-Accelerated Code Conversion

AI played a central role across the entire delivery lifecycle:

  • Code conversion: AI generated the majority of the converted code, with developers guiding and validating outputs
  • Complexity analysis: AI assessed application forms to estimate effort and prioritise delivery
  • Estimation: AI supported initial project sizing and ongoing refinement
  • Test generation: AI contributed to creating and validating tests during conversion

Developers acted as “human-in-the-loop”, setting direction, validating outputs, and ensuring quality, rather than manually rewriting the system.

3. Structured, Incremental Delivery

The system was migrated form by form, enabling:

  • Controlled progress through complex areas
  • Continuous validation and testing
  • Early identification of issues

The team worked in two-week sprints, maintaining familiar delivery practices while integrating AI into day-to-day workflows.

4. Engineering Guardrails and Quality Control

To ensure quality despite heavy AI usage, we implemented:

  • Test-led validation (aligned to TDD principles where possible)
  • Structured code reviews and oversight
  • Continuous testing to detect AI errors or “hallucinations”
  • Careful management of AI context and prompt design

We also developed approaches to handle AI limitations, such as:

  • Breaking work into smaller, well-defined chunks
  • Actively challenging AI outputs
  • Preventing unintended behaviour (e.g. deleting failing tests or incorrect code changes)

5. Knowledge Transfer and Ongoing Support

We reduced reliance on a single internal expert by:

  • Modernising the codebase into a more accessible and maintainable language
  • Involving a broader team in delivery
  • Setting up ongoing support and maintenance following delivery

Outcomes

50% Faster Delivery

The project was delivered at approximately half the time of a traditional approach, turning what was previously an 18+ month programme into a delivery measured in months.

Ahead of Schedule

The team progressed slightly ahead of plan, with strong indications the project would complete early despite aggressive initial timelines.

Improved Commercial Viability

AI significantly changed the economics of the project, making a previously unaffordable modernisation initiative viable within the client’s budget constraints.

Reduced Technology Risk

  • Eliminated reliance on unsupported VB6
  • Moved to a modern, supported C# .NET stack
  • Addressed key audit and compliance concerns

Preserved Business Continuity

Because the solution was delivered as a like-for-like conversion:

  • Users retained familiar workflows and interfaces
  • Disruption to operations was minimised
  • Adoption risk was significantly reduced

Foundation for Future Web Modernisation

The move to C# .NET WinForms created a clear pathway toward a future web-based architecture, without the risk and cost of attempting full transformation upfront.

New AI Delivery Capability

The project established a new delivery model where:

  • Developers are paired with AI rather than other developers
  • AI is used by default across engineering workflows
  • Teams continuously improve speed through rapid learning cycles

This resulted in significantly accelerated learning and delivery velocity over the course of the project.

Expanded Opportunity

The success of the engagement led to:

  • Further collaboration opportunities
  • Ongoing maintenance and evolution of the platform
  • Potential expansion of AI-enabled delivery approaches across other areas

What Made the Difference

This was not just a code conversion exercise. The value came from combining:

  • Deep modernisation expertise
  • Pragmatic architectural decision-making
  • Software craftsmanship principles
  • Hands-on experience with AI-driven delivery
  • The ability to balance speed with control and quality

While the client lacked the internal capacity and time to experiment with these approaches, Codurance brought proven practices and real-world experience to deliver both speed and confidence.

Key Takeaway

By combining AI-accelerated engineering with pragmatic modernisation, Codurance enabled the client to move off unsupported VB6 technology, reduce risk, and establish a modern foundation—delivering in months what would traditionally take years.

How Codurance can help

Navigating the transition from legacy systems to modern architecture is often a choice between high-risk overhauls or stagnating with technical debt. However, Codurance provides a third choice. By blending deep modernisation expertise with AI-accelerated delivery and software craftsmanship, we help organisations de-risk their business-critical applications in a fraction of the traditional time

Whether you're facing a looming compliance deadline or looking to reduce the cost of maintaining unsupported technology, we partner with your teams to deliver pragmatic, high-quality solutions that provide immediate value and a clear path for future growth. 

If you'd like to find out more, get in touch with us today. 

Frequently Asked Questions

1. Why choose a like-for-like migration to .NET WinForms instead of a full web rebuild?

A full rebuild as a web application was estimated to take over 18 months, which was commercially unviable. By performing a like-for-like migration to C# .NET WinForms, we eliminated the risks of unsupported legacy software and met strict audit deadlines in half the time, while providing a modern foundation for future web evolution.

2. How did AI accelerate the modernisation process?

AI was used as a force multiplier for our engineers. It handled the bulk of the code conversion, assisted in complexity analysis for better estimation, and supported test generation. This allowed our developers to act as high-level architects and validators, significantly increasing velocity without sacrificing quality. 

3. How did you ensure the quality of AI-generated code? 

We implemented strict "human-in-the-loop" engineering guardrails. This included test-led validation (TDD), structured code reviews, and breaking the migration into small, manageable chunks to prevent AI hallucinations. Our focus remained on software craftsmanship principles to ensure the new codebase was maintainable and robust.

4. Was there any disruption to the end-users during the migration?

The disruption to the end-users was minimal. Because the project focused on preserving existing workflows and UI elements, users could continue their core operational tasks in a familiar environment. The incremental, form-by-form delivery approach ensured that the system remained stable and validated throughout the entire process. 

About the Author

Natalie Gray is Director of Marketing & Growth at Codurance, a global AI-first software engineering consultancy that supports businesses and their investors to drive value through building and modernising sustainable software and platforms at all stages of the investment lifecycle. With more than 20 years in the tech industry, Natalie believes the power to innovate is accelerated when people are able to try new ideas, fail fast and collaborate. This is why, to her, community is at the heart of her approach to aligning business and technology outcomes. Outside of her commercial role, Natalie runs a number of tech meetups, is an advocate for women in tech and enjoys spending time with her family or protecting her 800 day streak on Duolingo.