The development world stands at a fundamental inflection point. Agentic coding (where AI agents autonomously write, test, and refine code) represents far more than another tool in the developer’s arsenal. It’s a paradigm shift that will redefine what it means to be a software developer.
Here’s what might surprise you: IBM i developers are uniquely positioned to thrive in this new era.
The Secret Weapon: Business Domain Knowledge
While much of the industry has spent years chasing the latest frameworks and languages, IBM i developers have been cultivating something far more valuable. They’ve been deeply embedded in their businesses, developing expertise in inventory systems, financial workflows, manufacturing processes, and customer lifecycle management that goes far beyond code.
This knowledge is becoming the most critical skill in software development.
Agentic coding tools are powerful, but they’re not magical. AI agents excel at translating clear requirements into working code, but they can’t divine what the business actually needs. They can’t understand the nuanced difference between how accounts payable should work versus how it currently works. They can’t spot the business logic edge case that will cause problems three months from now.
That’s where IBM i developers become indispensable.
The New Role: Architect of Requirements
In the agentic coding future, the developer’s primary value shifts from typing code to crafting precise business requirements and technical specifications. The better you can articulate what the system needs to do (including constraints, edge cases, and business rules), the better the AI agent will perform.
IBM i developers have spent decades translating complex business needs into RPGLE and COBOL. You’ve learned to think in terms of business processes, data flows, and system integration. You’ve debugged legacy code where the only documentation was the code itself, forcing you to reverse-engineer business logic from implementation.
These are exactly the skills needed to guide agentic coding systems effectively.
The Language Reality: Why RPG Struggles with Agentic AI Tools
However, there’s a critical technical reality we need to address. RPG simply doesn’t work as well with agentic coding tools. These AI systems are trained on vast repositories of modern code, primarily languages like Python, Node.js, Java, and C#. The training data, community resources, and established patterns that make AI agents effective are heavily weighted toward these languages.
This isn’t a criticism of RPG. It’s recognition that agentic coding tools need the ecosystem advantages that come with widely adopted modern languages. When you give AI the right setup, it can write, test, and validate code faster and more accurately than ever before. But that “right setup” requires modern languages with robust AI tooling support.
This is where transformation becomes essential – not just desirable, but necessary for leveraging the future of development.
The Testing Revolution
Another critical component of successful agentic coding is automated testing. For AI agents to work independently and iterate rapidly, they need to test their own changes without humans in the loop. This creates a feedback loop where the agent can verify its work, catch errors, and refine its approach autonomously.
This requires robust testing infrastructure with comprehensive coverage: unit tests, integration tests, and automated validation of business rules. The better your testing framework, the more confidently your AI agent can operate.
For many IBM i shops, this represents a significant opportunity. Traditional development practices often relied on manual testing and deep institutional knowledge. The shift to agentic coding demands a more systematic approach to quality assurance.
The Infrastructure Foundation
Making agentic coding work requires foundational changes to how we manage code. Two critical pieces of infrastructure need to be in place:
Git for Source Control
Modern AI development tools are built around Git workflows. Moving from traditional IBM i source control to Git enables better collaboration with AI agents, version tracking, and integration with the broader development ecosystem.
CI/CD Pipelines
Continuous integration and continuous deployment processes are essential for agentic coding. They provide the automated build, test, and deployment cycles that allow AI agents to iterate quickly and safely.
These aren’t just prerequisites for agentic coding. They’re investments that pay dividends even before you fully embrace AI-driven development.
Beyond Language: The Modern Environment Advantage
Language transformation alone isn’t the complete answer. The real advantage comes when modern code runs inside a modern environment with automated testing, CI/CD, agentic AI tools, and modular architectures that make innovation faster, safer, and future-ready.
This is the complete package that positions your development team to leverage agentic coding effectively. It’s about creating an ecosystem where AI agents can operate with confidence, where business logic is preserved and enhanced, and where your deep domain expertise becomes the guiding force behind autonomous development.
How Profound AppEvo Bridges the Gap
For years we’ve been working through these exact challenges with RPGLE and COBOL shops. Our approach recognizes that the path to agentic coding doesn’t require abandoning your existing systems or enduring risky “big bang” rewrites.
Profound AppEvo uses specialized Algorithmic + AI Refactoring, a hybrid method that combines the reliability of deterministic algorithms with the power and adaptability of AI.
This approach delivers:
- Predictable transformation of your core business logic
- Preserved business rules accumulated over decades
- Modern language output (Node.js, Python, Java, C#) that works seamlessly with agentic coding tools
- Incremental futurization without full system downtime
- Infrastructure modernization including Git integration and CI/CD pipelines
Traditional modernization tools rely on manual rewrites or risky AI-only conversions. Our hybrid approach ensures your systems evolve safely, efficiently, and intelligently, preparing the foundation for agentic coding while maintaining business continuity.
We help you build the right bridges while maintaining the stability your business depends on. The transformation includes not just code conversion, but the complete infrastructure setup that makes agentic coding possible: automated testing frameworks, modern deployment pipelines, and modular architectures.
The Path Forward
The rise of agentic coding doesn’t diminish the role of the developer. It elevates it. Instead of spending time on syntax and boilerplate, you’ll focus on what you’ve always done best: understanding the business and translating that understanding into precise specifications.
Your decades of business domain expertise aren’t becoming obsolete. They’re becoming more valuable than ever.
The organizations that will thrive in the agentic coding era are those that combine deep business knowledge with modern development infrastructure. For IBM i shops, that means leveraging your unparalleled understanding of business processes while transforming your development practices to support AI-assisted coding.
The future of development isn’t about replacing developers with AI. It’s about empowering developers with AI. And IBM i developers, with your deep business knowledge and commitment to reliable systems, are perfectly positioned to lead this transformation—once you have the right foundation in place.
Ready to Position Your Team for the Agentic Future?
The infrastructure changes required for agentic coding (Git integration, CI/CD pipelines, automated testing frameworks, modern languages) can feel daunting, especially in established IBM i environments where these patterns may be new.
We’re happy to share what we’ve learned about building these foundations while preserving your critical business logic and maintaining operational stability.
Ready to take the first step? Reach out to our team of experts at futurization@profoundlogic.com to begin positioning your team for the agentic coding future.