Red Hat Developing AI Tool "Log Detective" To Help Developers

Jiri Kyjovsky of Red Hat has shared news today of Log Detective, a new tool being developed that will leverage an AI model to help in analyzing build failures for RPM packages.

The developers are currently working on Log Detective and training an AI model to "recognize and explain build failures in simple words." Log Detective aims to make it easier for RPM packagers to figure out why a build may have failed. Their new project website argues:

"Each build produces thousands of lines of output split among multiple log files. And the relevant error message can be anywhere. It's just like a needle in a haystack.
....
Veteran packagers have an intuition where the error message will most likely be, but the process is tedious regardless. Newbies are often overwhelmed by the complexity and miss the error message completely.
...
Train an AI model to understand RPM build logs and explain the failure in simple words, with recommendations how to fix it. You won't need to open the logs at all."

Well, that's how they hope Log Detective will work out at least. Currently they are working on collecting more RPM build logs still for training their AI model.

Log Detective was announced a short time ago on the Fedora devel list. Their current basic website for Log Detective can be found at log-detective.com. It will be interesting to see eventually how this works out in practice and if the AI model can be extended to help with analyzing other Linux build logs, etc.