AI-assisted development is moving pretty quickly, but I’m curious how other teams are handling the practical side of it.
We’re seeing more teams use AI for coding, documentation, ticket analysis, testing, and repetitive engineering tasks, but the challenge seems to be keeping everything traceable through the existing Jira workflow.
For example:
Are you using AI to create or refine Jira issues?
How are you reviewing AI-generated code before it moves through the workflow?
Are you connecting AI agents to Jira Automation?
What do you do when an AI-generated change needs human approval?
Have you found a good way to measure whether AI is actually improving delivery rather than just increasing activity?
I’ve been looking at how engineering teams approach this from a production perspective, and Software Developers Pro has also been exploring the broader shift toward AI-assisted engineering rather than treating AI as a separate tool.
Curious what has actually worked for teams here, especially in larger Jira environments. Are you keeping AI mostly outside the workflow, or starting to make it part of the development process?
Hi @Wyatt Lewis ,
I cannot comment much from the coding perspective, as I'm not a software developer (although I do have some personal/side projects), but when it comes to my experience and what I've seen around:
Are you using AI to create or refine Jira issues?
Yeah! I create almost all work items via Rovo or MCP connectors. You do need to explain it as detailed as possible your 'way of thinking' - but AI does a pretty good job when it comes to this 🙂
Are you connecting AI agents to Jira Automation?
Depends on the case, but yes - we're trying to 'extend automation' with AI capabilities. Note that it might not be as stable as automations (as with automations, you know what you'll get), but it's still a good way to see what's possible and what's not when it comes to embedding agents in automation flows. 👀
Have you found a good way to measure whether AI is actually improving delivery rather than just increasing activity?
Haven't got to that part yet 😅 But from my experience, and that's mainly from creating new work items, time-saving is HUGE. And I would probably spend way too much time thinking about how to structure or describe a work item.
Yet, some metrics or insights would be nice.
From the consultancy perspective, I've also been exploring things like AI-generated workflows and other CIs. Creation from scratch, or even just reviewing them, could be improved with the new capabilities of AI tools.
Cheers,
Tobi
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.