JupyterHub is an important tool for research and data-driven decisions at Globo.com. Diogo Munaro Vieira and Felipe Ferreira explain how data scientists at Globo.com—the largest media group in Latin America and second largest television group in the world—use Jupyter notebooks for data analysis and machine learning, making decisions that impact 50 million users per month.
By using Jupyter notebooks combined with JupyterHub, Globo.com’s data scientists can analyze company products metrics, recommendation systems results, and A/B test results and build their own machine learning models. All of the studies support OAuth 2, Spark jobs monitoring, Python, R, PySpark, and SparkR without installation and configuration issues. Globo.com’s data science platform empowers data scientists and managers. Running on two machines, each with 32 CPU cores and 125 GB of memory, the platform is supported by JupyterHub and is used to filter billions of metric events a day.
Diogo Munaro Vieira is a big data engineer at Globo.com. He is experienced in web development, P2P networking, collaborative systems, recommendation systems, open source software, and business intelligence. Diogo holds a bachelor’s degree in biological science and bioinformatics and a master’s degree in artificial intelligence from Universidade Federal do Rio de Janeiro.
Felipe Ferreira is a big data engineer at Globo.com, where he focuses on big data platform analytics using Hadoop and associated ecosystem tools. Felipe is an analytical, performance-focused engineer with over 12 years of experience in enterprise systems development and architectural design using JEE technology combined with the ability to drive user-centric solutions, define strategy, and lead data management.
Comments on this page are now closed.
©2017, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org