Using Static Code Analysis to Improve Java APIs

In this blog, I detail my experience in utilizing various static code analysis tools as part of the build process for Sports API.

GumGum recently hosted Dr. Genquan Stone Duan and Andrew Pierno of WiZR at our LA Computer Vision Meetup. This presentation is an interactive exploration of WiZR's machine learning and computer vision infrastructure, which is used to provide real-time analytics for the purposes of security and surveillance.  Anyone interested in embedded machine learning and computer vision should check out this technical deep-dive. 

Ephemeral Development Environments

At GumGum we build alot of in-house web products. The development for these products follow a dev/stage/prod pipeline, but often times a single development environment is not enough. This post describes our migration to using traefik and drone to build out dynamically ephemeral preview development environments for our web products.

Code Duplication & Orthogonality -- eliminate one and focus on the other. In this episode of The Cat Ate My Source Code, Azam, Ben, and Collin cover the evils and benefits of these topics drawing their ideas from The Pragmatic Programmer and their experiences so far.


On Monday, April 23, GumGum held our first Women Who Lunch event. This recurring event was started to bring women from the tech industry into our office to share their story and answer questions from the women in tech here at GumGum.

K Means Clustering for Sports Images

GumGum Sports processes enormous amounts of media each day. They come from a variety of sources and forms, including social media posts and broadcast/streaming videos. Our goal is to identify media that is relevant to our clients to estimate the value of their sponsorships and placements

React 16's Stellar New Portal API

React 16 landed with a helpful new API called portals, which is a first-class way of rendering children into a DOM node outside of the parent component's hierarchy. We'll demonstrate how the portal API can simplify the relocation or transportation of DOM markup without resorting to prop functions, Redux, or other libraries with some fun example apps.

Sanja Stegerer made a presentation about determining brand safety using Natural Language Processing at Women in Software Engineering meetup recently.

We are excited to announce a new software engineering podcast, The Cat Ate My Source Code.

Maxime Nay, Lead Data Engineer at GumGum gave a talk explanaing GumGum's Data Architecture and challanges associated with it on February 15th, 2018 at South Bay Java User's Group.

Processing 100 Billion Events a Day from GumGum on Vimeo.


Over the last 10 years or so we have seen an exponential evolution of the Internet experience, we went from the linear desktop-based low-bandwidth kind-of-ugly interfaces to the modern responsive video-driven asynchronous Web that we all love nowadays. This happened due multiple factors but importantly because of the popularization of JavaScript as the de facto language of the Web Browser.

Cassandra to DynamoDB: Better performance and cost benefits

We recently  moved to Amazon DynamoDB from Apache Cassandra. In this blog I will be discussing the architecture, design decisions made in the process with their justifications and the steps taken in order to complete this switch.

Text extraction using Dragnet and Diffbot
Amazon Redshift Spectrum: Extending Our Data Warehouse Capabilities

Amazon Redshift is the primary data warehousing solution used at GumGum. Apart from the real time reports which are powered through Druid, Redshift fuels the majority of our reporting capabilities. Being a fully managed solution from Amazon, we do not have to maintain the Redshift cluster, be it the hardware or the redshift engine. Redshift is column oriented and its massively parallel processing architecture (MPP) makes it petabyte scalable. Amazon Redshift also includes Redshift Spectrum which can directly query unstructured data stored in S3.