In a recent blog post, Nanyu Chen, Senior Applied Research Engineer at LinkedIn, has provided an overview of a new, comprehensive technical paper on the XLNT Platform, LinkedIn's A/B testing platform.
XLNT was built by LinkedIn to help the company make data-driven A/B testing decisions. XLNT was designed to encompass three steps of the testing process: design, deploy and analyze.
As Chen points out in his
blog post, LinkedIn is an organization running hundreds of experiments daily with interactions posing a serious threat to experiment trustworthiness. The company uses XLNT to address its most common concerns and use cases related to interactions between experiments.
I strongly suggest reading his blog post prior to diving in the technical paper. The paper itself is very in-depth and scholarly in nature and the blog post does a great job setting up the paper’s offerings in a more easy to digest way. But if you’re determined to do so, you can dive right into the paper itself. LinkedIn provides direct, ungated
access to the 10 page document.