Inspiration
We are both Software engineers and we spend lot of time triaging bugs, production incidents and other kind of issues. We are also very lazy and would really appreciate having a tool making our life easier.
What it does ?
- Tom: Kelly, can you please look at this error ?
- Kelly : 'Alice could not log in ?' Again ? ... It's - the fifth user this week.
- Tom: Oh really ? Is this issue tracked on Jira ?
- Kelly: ... nope didn't really had the time to create it.
...
And here comes the Smart Jira Tracker 🎉🎉🎉
In this situation Kelly could have connected our App to her logging system and automatically it would have appeared on Jira as an issue. The app being smart, it will be able to handle Kelly's situation by gathering issues into a single one and displaying the number of occurrences.
How we built it
Using Forge.
Challenges we ran into
We really wanted to make the bot as smart as possible by plugging it in with a machine learning API. Google AutoML looked to be really well suited to us. However we ran out of time and decided not to add it as part of the submission.
Accomplishments that we're proud of
We have developed an intelligent algorithm to compare events.
What we learned
We really enjoyed working with Forge, it is an amazing platform to work with.
What's next for Smart Jira Tracker
We really hope to get the jury attention and why not a blog article. If we see people are interested, we will definitely make it smarter by focusing on the machine learning part. We want it to react to all sort of events, acting on its own and/or provide suggestions such as:
- Assign automatically an issue to someone based on previous activity of that user.
- Closing bugs no longer appearing in production after a certain time.
- Someone creates an issue, the app detects similarity with an existing one, provides a suggestion: "it looks really similar to issue XXXX, we advise to close it" ...
Log in or sign up for Devpost to join the conversation.