Hey Indie Hackers ✌️
Defects AI is a service to predict labels for GitHub Issues fully automated, with machine learning models trained in particular to that use case. Don’t waste your time by labeling Issues, use your time to solve them.
How did everything start?
Defects AI began as a side project, a statistics tool to provide you with Insights about your product quality. While digging deep into different datasets, I figured many GitHub Issues are not labeled.
After a couple of days and talks later I realized why: For many projects, it is difficult to analyze issues in an acceptable amount of time.
That was the turning point, I paused working on the statistics tool and started working on the automated issue labeling for GitHub. Starting with the core feature, the machine learning model to predict the GitHub issue type.
A few successful first tests and further improvements of the machine learning model gave me the confidence to continue that idea.
How does it work?
The setup is simplistic, register at Defects AI, install the GitHub Defects AI App and select the repositories which you want to enable for the automatic issue labeling.
Each time a new issue is created, GitHub sends a request to our server, we predict the issue type and send it back to your repository. The label is set automatically to the issue.
Automatic Issue Labeling in action
How did I manage to build Defects AI while working full-time?
- Reduce the features to the bare minimum
- Save your weekends for your family and relax
- I usually worked 5 days a week on Defects AI, each day at least 1 hour before my actual job starts.
(You are still responsible do deliver great results to your actual employer. It is up to you to manage both. If you aren’t capable of combining both please don’t risk your job)
I spend a bit more than 3 months (95 Days) to build the initial release of Defects AI, to be exact I started on the 15–06–2018. Here is the first commit to the repository:
What’s next?
While collecting feedback from our user we continue working on Insights and improving our models further. Feel free to follow us on Twitter.
We continuously improve our models, supported currently: Bug, Feature, Documentation, and Question.
A few words about money
Like everything else also a SaaS costs money, the biggest spendings are the training costs for our models hosting and collaboration tools. We spend around 12$ for each training, as you can imagine in the beginning we trained a lot.
We plan to publish our financial reports each month.
Curious?
We provide you with a simple and easy way to experiment with our prediction service, just go to our website and paste some of your issues:
https://defects.ai/feature/automatic-issue-labeling
❤ Thank you for having you here. Leave me some comments if you have questions about building a SaaS.
I appreciate your feedback to get inspired for my next posts.
Cheers
Sascha