Put open source to work
July 16–17, 2018: Training & Tutorials
July 18–19, 2018: Conference
Portland, OR

Applying optical character recognition and Kubernetes to Twitch

2:35pm3:15pm Thursday, July 19, 2018
Artificial intelligence, Kubernetes
Location: D137/138
Level: Beginner

Who is this presentation for?

  • Developers, distributed computing enthusiasts, and gamers

Prerequisite knowledge

  • A basic understanding of Kubernetes and microservice architectures

What you'll learn

  • Explore the open source application Rotisserie
  • Learn how it uses a number of technologies, including a Node.js engine, a microservice architecture, and the open source optical character recognition software Tesseract in a microservice


Cullen Taylor offers an overview of Rotisserie—an open source Node.js application that runs on Kubernetes—which applies the concept of the red zone in American football to the popular online battle royal game PlayerUnknown’s Battlegrounds (PUBG) with the goal of always viewing the most popular PUBG Twitch stream with the least amount of people alive in game, as this is the point where streamers are playing at their highest levels.

Cullen explains how Rotisserie was built from open source technologies, including a ton of Node.js libraries, Tesseract, Docker, and more. The Rotisserie team at IBM is continuously developing the technology to include a machine learning model using TensorFlow, and the frontend has recently been overhauled to a React-based dashboard. Cullen shares plans for future work and explains how you can jump on board.

Photo of Cullen Taylor

Cullen Taylor


Cullen Taylor is a developer advocate at IBM. Previously, he was a DevOps engineer at IBM.