Skip to main content

GitLab acquires UnReview as it looks to bring more ML tools to its platform

DevOps platform GitLab today announced that it has acquired UnReview, a machine learning-based tool that helps software teams recommend the best reviewers for when developers want to check in their latest code. GitLab, which is looking to bring more of these machine learning capabilities to its platform, will integrate UnReview’s capabilities into its own code […]

DevOps platform GitLab today announced that it has acquired UnReview, a machine learning-based tool that helps software teams recommend the best reviewers for when developers want to check in their latest code. GitLab, which is looking to bring more of these machine learning capabilities to its platform, will integrate UnReview’s capabilities into its own code review workflow. The two companies did not disclose the price of the acquisition.

“Last year we decided that the future of DevOps includes ML/AI, both within the DevOps lifecycle as well as the growth of adoption of ML/AI with our customers,” David DeSanto, GitLab’s senior director, Product Management – Dev & Sec, told me. He noted that when GitLab recently surveyed its customers, 75% of the teams said they are already using AI/ML. The company started by adding a bot to the platform that can automatically label issues, which then led to the team meeting with UnReview and, finally, acquiring it.

Image Credits: GitLab

“Our primary focus for the second half of this year in bringing on UnReview is to help automate the selection of code reviewers. It’s a very interesting problem to solve, even we at GitLab occasionally end up picking the wrong reviewers based off of what people know,” DeSanto noted.

GitLab launched its original code review components last year. As Wayne Haber, GitLab’s director of Engineering, noted, that was still a very manual process. Even with the new system, teams still retain full control over which reviewers will be assigned to a merge request, but the tool will automatically — and transparently — rank potential reviewers based on who the system believes is best suited to this task.

“I am grateful for the opportunity to share my passion for data science and machine learning with GitLab and its community,” said Alexander Chueshev, UnReview’s founder (and now a senior full stack engineer at GitLab). “I look forward to enhancing the user experience by playing a role in integrating UnReview into the GitLab platform and extending machine learning and artificial intelligence into additional DevOps stages in the future.”

DeSanto noted that GitLab now has quite a bit of experience in acquiring companies and integrating them into its stack. “We’re always looking to acquire strong teams and strong concepts that can help accelerate our roadmap or strategy or help the platform in general,” he said. “And you can see it over the last couple of years of acquisitions. When we were looking at extending what we did in security, we acquired two leaders in the security space to help build that portfolio out. And that’s fully integrated today. […] In the case of this, UnReview is doing something that we thought we may need to do in the future. They had already built it, they were able to show the value of it, and it became a good partnership between the two companies, which then led to this acquisition.”

One interesting wrinkle here is that GitLab offers both a hosted SaaS service and allows users to run their own on-premises systems as well. Running an ML service like UnReview on-premises isn’t necessarily something that most businesses are equipped to do, so at first, UnReview will be integrated with the SaaS service. The team is still looking at how to best bring it to its self-hosted user base, including a hybrid model.

GitLab’s head of Remote on hiring, onboarding and why Slack is a no-work zone

Data & News supplied by www.cloudquote.io
Stock quotes supplied by Barchart
Quotes delayed at least 20 minutes.
By accessing this page, you agree to the following
Privacy Policy and Terms and Conditions.