Twitter Engineering

Building the world’s first real-time distributed public platform connecting people to their interests, events and each other

From the Twitter Engineering Blog

  • Behind the scenes of enhancements to MoPub data Tuesday, November 17, 2015

    As a leading monetization solution for mobile app publishers, we’re excited to share some news about major infrastructure improvements on MoPub. We’ve been listening to the developer community and based on feedback we’ve received over the months, we’ve made MoPub better by equipping publishers with faster access to data and faster updates between our UI and ad server.

  • Evaluating language identification performance Monday, November 16, 2015

    Understanding the content of Tweets is important for many reasons: grasping a user’s interests (which in turn lets us show more relevant content), improving search, and fighting spam. There are many steps involved in a typical natural language processing pipeline, but one of the first and most fundamental steps is language identification — determining the language in which a piece of text is written.

  • Autograd for Torch Friday, November 06, 2015

    We recently established the Twitter Cortex team with a simple mission: to build a representation layer for Twitter in order to improve our understanding of our content and users.

  • Twitter experimentation: technical overview Friday, November 06, 2015

    In our previous post, we discussed the motivation for doing A/B testing at Twitter, and how A/B testing helps us innovate. We will now describe how the backend of Twitter’s A/B system is implemented.