In part 1, we described why 2025 marked a turning point for AI in software development and how AI became productive once it was deeply connected to real-world systems and frameworks.
Tooling as the Real Accelerator
By early 2024, we began using Windsurf - at the time still known as Codeium - as an AI-assisted development tool.
Even without formal OpenEdge or ABL support, the results were often astonishing. This tool did not merely autocomplete code; it detected logical flaws, suggested improvements, and helped clean up existing implementations.
Key insight: Modern AI thrives on context.
Tooling that provides access to source code, coding standards, architectural documentation, and historical patterns dramatically increases AI effectiveness.
Why VS Code Changed the Equation
VS Code emerged as the ideal foundation for this new way of working.
Unlike Eclipse-based environments, it decouples the development experience from specific OpenEdge versions. Whether working with OpenEdge 11.7, 12.2, 12.8, or upcoming 13.0 releases, developers benefit from the same AI-driven workflow.
At the same time, the AI tooling landscape diversified rapidly.
IDE-integrated tools such as Windsurf, Cursor, Copilot, and Kilo-Code were complemented by:
Beyond IDE integrations, the ecosystem expanded quickly:
- CLI-based tools such as Claude Code
- Cloud-native agents like Devin
These tools go far beyond code generation - they analyze entire codebases, answer architectural questions, and assist with refactoring at scale.
Source Control as an AI Enabler
One of the most impactful - and initially underestimated - consequences of AI adoption was its direct impact on Consultingwerk’s source control strategy.
AI tools overwhelmingly target GitHub first. Pull requests, agent-based branches, and parallel task execution all assume a GitHub-centric workflow.
This realization led us to rethink long-standing SCM decisions.
While Perforce and Bitbucket offered strengths we valued, GitHub’s integration depth with AI tools ultimately proved decisive.
Why This Matters
OpenEdge teams that historically placed little emphasis on SCM need to recognize a hard truth:
AI cannot be successfully adopted without a strong source control strategy.
At Consultingwerk, this realization fundamentally reshaped how we reassessed tooling, workflows, and long-standing assumptions around version control.
Our DevOps team supports SCM adoption in every scenario - from small legacy codebases and file-based workflows to complex, distributed enterprise environments - enabling teams to adopt AI-ready processes pragmatically and sustainably.
Outlook to Part 3
Tools alone, however, do not stand still. In the next part, we explore why constant change has become the new normal - and how to stay productive in an environment of continuous evolution.