From Assisted to Autonomous: A DevOps Series for IBM i
AI coding tools are only as good as the environment they’re working in. This two-part series covers what it takes to get IBM i development ready for AI, from Git and parallel development to the documentation and testing that let AI move from assisting your team to working on its own.
Each session has its own registration. Click the title of each webinar below to expand its description and registration box, and register for both sessions separately to attend the full series.
Getting to AI-Assisted Development: Git and Parallel Development for IBM i
AI coding agents can generate a lot of code fast, sometimes touching dozens of objects in a single request. That speed runs headfirst into a habit a lot of IBM i shops still have: locking a source member the moment one person opens it, so nobody else can touch it until they’re done.
That restriction doesn’t just slow down human developers anymore. It actively blocks AI agents from doing their job. This session covers why Git-based DevOps and real parallel development are the first requirement for any AI-assisted IBM i workflow, not a nice extra to add later. You’ll see what it takes to move off the one-person, one-member model, and what opens up once your team, human or AI, can finally work on the same codebase at the same time.
If you would like to register for session two as well, please scroll down and click the title of the webinar to expand.
From Assisted to Autonomous: The Next Step in AI Development for IBM i
Getting your code into Git solves one problem. It doesn’t solve the next one: AI still needs compile/build information, object relationship documentation, test specifications, and change documentation to do anything more than just generate code. With access to that information, the AI code assistant can write code, create objects, fix compile errors, identify contextual information from related objects and documentation, and run tests on the generated code without intervention. It is where the real value from AI assisted coding comes from.
This session picks up where the Git foundation leaves off. We’ll cover what it takes to give AI tools accurate, structured feedback, the kind that lets them catch a problem before it reaches production instead of after. You’ll walk away with a clearer picture of what “AI-ready” really requires once you’re past version control and why skipping this layer is the reason a lot of AI coding experiments stall out before they deliver anything useful.