You're on your way to the next level! Join the Kudos program to earn points and save your progress.
Level 1: Seed
25 / 150 points
1 badge earned
Challenges come and go, but your rewards stay with you. Do more to earn more!
What goes around comes around! Share the love by gifting kudos to your peers.
Keep earning points to reach the top of the leaderboard. It resets every quarter so you always have a chance!
Join now to unlock these features and more
In this article, our focus is on addressing these challenges and introducing an AI-powered assistant designed to enhance your backlog refinement process, potentially making it more efficient and enabling valuable collaboration.
Drawing from our firsthand experience, backlog refinements can pose significant challenges, often stretching over time. As a product owner, communicating everything comprehensively can be a daunting task. This may be due to being new to the role, dealing with stories that were shelved as loose ideas in the backlog and have now gained sudden importance, or even just having a bad day where writing acceptance criteria becomes challenging.
Hence, we crafted a solution to address this issue. With RefineMeFaster, our AI-driven Jira app, we aim to tackle this task
Our app streamlines the process of improving user stories in Jira. The app analyses issue data like type, summary and description and tries to find unclear sentences and gives you hints how to improve them. Those suggestions can be accepted, rejected or marked as resolved.
The field edit dialog with improvement suggestions
The "Improvements" field is simply another custom field. You can create it through the familiar admin section called "Custom fields" and add it to the screens where it seems appropriate.
Create the Improvements field as every other custom field
Once you've included this field in a screen, go to an issue and open the custom field dialog to initiate the improvement process. At this stage, review and prepare the data for analysis, ensuring the removal of sensitive information.
AI-powered analysis is employed to generate improvement suggestions by examining this data.
The choice to keep suggestions open (indicating they are unresolved but valid) or discard them is entirely up to you. The number of open suggestions will be displayed as a numeric value within the custom field after saving, aiding you in monitoring ongoing improvements.
You can also mark suggested improvements as resolved if you make changes to the sentences. Furthermore, the open suggestion count supports Jira Query Language operations, such as sorting issues based on the highest number.
The color changes in accordance with the open suggestion count
In addition, it can be integrated into automation, enabling the triggering of actions or workflows when the number of open suggestions surpasses a predetermined threshold.
The application emerged during Atlassian Codegeist Unleashed 2023, and we had a delightful time combining Jira (Forge) with an AI service (ChatGPT). Although we had some familiarity with Forge, our initial challenge was implementing the AI component.
We began with a relatively straightforward task to establish a proof-of-concept: generating release notes from issue data. To accomplish this, we introduced a custom field that allows the creation of a changelog per issue. We believe that having this field as close as possible to the issue view makes perfect sense, making it an ideal choice for a custom field.
Another crucial concern pertains to transparency and privacy. As a response, we've incorporated an admin page that provides details about the AI services in operation, along with the prompt and data that are transmitted to the AI.
Find important information about the app and used AI services
If you're interested in delving into the technical intricacies, please feel free to explore our Codegeist submission.
Wanna try it out? Here's the install link