The operators of the version management platform GITLAB have announced the overview of unreaview. The technology behind uses Machine Learning to seek software development projects and, for example, to propose the most suitable review for code reviews. According to Gitlab-CTO Eric Johnson, the acquisition marks the first step towards the integration of "Applied Machine Learning for Devops" on the platform.
Find reviewer faster
UNREVIEW collects the information from Pull Requests (PR) and Commits and Participated by Pull Requests (PR) and Commits and participating authors and reviewers. The data thus determined cleans up the system, it leads together and stores it for the analysis cluttered by machine learning. On the basis of the trained ML models, developers can then determine recommendations for reviewers, for example based on PRS from their project. The network graph provided by Unreview allows the proposal with the actual project data.
The fact that the use of artificial intelligence or machine learning in the DEV (SEC) OPS environment, especially as a foundation stone for an automated testing of software is always stronger in the foreground, is one of the essential insights from the gitlabs DevSecops Survey 2021. Accordingly, the deployment cycles of source code doubled in the observation period – also fueled by the pandemic. Stronger automation in vital phases of software development is therefore unprecedentable.
Automation of the DEVSECOPS process chain with ML
Gitlab drives this process with "Applied Machine Learning for Devops" Targeted forward, in whose framework uncieved especially the DEV phases "Administer" (Manage), "To plan" (Plan) and "Create" (Create) should further optimize. In addition, further areas in the DevoPS process chain are to be developed – for example, to pave data teams the way from Dataops over Mlops to Modelops.
Further information on the overview and the GITLAB strategy can be found in official envision. The unreaview technique should be fully integrated by the end of the year in Gitlabs’ SaaS version of Code Review. Details about the progress of the integration can be tracked up to then in the associated GITLAB EPIC.