When business leaders think about AI, their minds often jump to customer-facing chatbots, predictive analytics or flashy new digital products. But some of the biggest and fastest ROI opportunities from AI are not visible to customers at all. They are buried deep inside existing codebases.
In a recent episode of Staying Competitive, hosts Ray Beste and John Eisenhauer were joined by Russ Tolsma, Director at Sikich, to explore one of the most practical and valuable applications of AI today: AI-enabled code transformation. The conversation highlighted a clear reality for executives and technology leaders. Modernizing how business apps are built, maintained, and evolved is becoming central to data and AI solutions that actually drive business outcomes.
Check out the full episode on Spotify or iHeart Radio!
What is AI-enabled code transformation?
AI-enabled code transformation focuses on using modern data and AI services to improve the entire software development lifecycle, not just code creation. It goes well beyond writing code faster.
This approach includes:
- Refactoring legacy code to improve flexibility and maintainability
- Automating steps across the development lifecycle, from requirements through deployment
- Using AI agents to analyze, document and test existing systems
- Reducing friction between business intent and technical execution
Many organizations rely on internal applications that were built years ago or even decades ago. These systems still function, but they are often slow to adapt, costly to maintain and difficult to understand. AI changes that equation by making those systems easier to evolve instead of replace.
The real cost of technical debt and why it matters now
Executives often hear the term “technical debt” and ask a reasonable question. If this has existed for years, why address it now?
The answer is not found on a balance sheet. It shows up in time to value. Technical debt becomes visible when:
- New features take longer than expected to deliver
- Teams struggle to pivot as markets change
- Small updates introduce unexpected failures
- Engineering teams spend more time maintaining systems than innovating
Data and AI-enabled business solutions shorten the distance between a business idea and a validated outcome. They do not replace architectural discipline or human judgment, but they reduce friction that slows organizations down.
Beyond autocomplete: the rise of AI agents in development
Most leaders are familiar with AI tools that help developers autocomplete code or generate snippets. That is only the beginning.
We are now seeing the emergence of AI Agents in business apps, systems that operate continuously in the background inside tools teams already use such as source control systems and issue trackers. These agents can:
- Monitor repositories and issue trackers
- Evaluate requirements as they are created
- Reduce manual effort for tasks developers
- Generate documentation and test cases automatically
- Support developers without replacing human decision-making
For business stakeholders, this matters because requirements and intent can flow more directly into execution, reducing misalignment and rework.
Can AI refactor and generate production-ready code?
The short answer is not autonomously… yet.
“Vibe coding” and one-shot application tools are impressive and increasingly accessible.
They are well suited for:
- Proofs of concept
- Rapid prototypes
- Internal tools
- Early exploration of new ideas
However, production-grade systems still require governance, security compliance and architectural oversight. These areas remain critical for regulated industries and customer-facing platforms.
Today’s tools can move teams most of the way there, but experienced engineers are still essential to ensure quality and resilience. The real risk is not AI-generated code. The risk is deploying code without appropriate guardrails.
Refactoring explained for executives
Over time, software accumulates decisions that make sense at the moment but restricts flexibility later. When organizations need to pivot due to new regulations, customer expectations or competitive pressure, those earlier decisions slow progress.
Refactoring helps organizations:
- Adapt more quickly to market changes
- Reduce the risk of breaking existing functionality
- Deliver new capabilities faster and at lower cost
AI accelerates refactoring by analyzing undocumented systems, explaining how they work and generating updated documentation and tests.
Testing documentation and the work developers avoid
Some of the most meaningful benefits of AI come from automating work that development teams traditionally avoid or postpone:
- Writing test cases
- Maintaining documentation
- Onboarding new engineers
AI agents can analyze requirements and existing code to generate comprehensive test coverage and documentation automatically. This makes test-driven development more achievable and improves reliability. The result is faster releases, fewer surprises and greater confidence when systems change.
Who benefits most from AI code transformation?
While nearly every organization can benefit, early gains are often strongest for:
- Small high-impact engineering teams
- Businesses operating in fast-moving or competitive markets
- Organizations with aging but mission-critical systems
Larger enterprises can also benefit, but adoption requires thoughtful change management and phased implementation. The goal is not to transform everything at once but to focus on areas where AI can unlock the most value.
How to get started without disruption
AI-enabled code transformation is not a rip-and-replace effort.
A practical approach includes:
- Identifying high-friction areas in the codebase
- Piloting AI tools for documentation testing or refactoring
- Measuring impact and adoption
- Expanding where results justify investment
This mirrors how organizations successfully adopt other modern data and AI services. Right-sizing the approach is essential.
Final thought: a strategic imperative
AI-enabled code transformation is not just a technical improvement. It is a strategic capability that supports faster adaptation, stronger execution, and ultimately driving exponential growth.
Organizations that thoughtfully apply data and AI solutions to their existing business apps and systems will move faster and compete more effectively than those that treat AI as only a customer-facing experiment.
Stay ahead of what is next. Subscribe to the Staying Competitive podcast on Spotify or iHeart to hear practical insights from industry leaders and experts shaping modern data and AI solutions.
This publication contains general information only and Sikich is not, by means of this publication, rendering accounting, business, financial, investment, legal, tax, or any other professional advice or services. This publication is not a substitute for such professional advice or services, nor should you use it as a basis for any decision, action or omission that may affect you or your business. Before making any decision, taking any action or omitting an action that may affect you or your business, you should consult a qualified professional advisor. In addition, this publication may contain certain content generated by an artificial intelligence (AI) language model. You acknowledge that Sikich shall not be responsible for any loss sustained by you or any person who relies on this publication.
