Data Scientist - Fraud Detection
AdColony is a mobile video advertising company whose proprietary Instant-Play™ technology serves razor-sharp, full-screen video ads instantly in HD across its network of iOS and Android apps, eliminating the biggest pain points in mobile video advertising: long load times and grainy, choppy video.
As a leading mobile video advertising and monetization platform, AdColony works with both Fortune 500 brands and more than half of the top grossing publishers in the App Store. The company’s reach, targeting and optimization tools and services provide advertisers with a superior way to engage mobile audiences at scale. AdColony’s app developer tools and services provide publishing partners with ways to maximize monetization while gaining insight needed to continuously optimize content and advertising offerings.
AdColony is a division of Opera Mediaworks and has offices in Los Angeles, San Francisco, Seattle, Chicago, New York, and the UK. AdColony was recently listed in the 2013 and 2014 Inc. 500 (Fastest Growing Private Companies in America) and named Best Mobile Ad Network by Digiday.
AdColony is an equal opportunity employer. For more information, please visit www.adcolony.com
Your mission is to design algorithms to detect, prevent and predict fraudulent activities in mobile advertising. You will also have the opportunity to build predictive models to help improve efficiency and relevancy on mobile advertising for hundreds of millions of users worldwide.
If you have experience in fraud detection, and have deep understanding of statistics and machine learning algorithms, we want to talk to you.
This position will be located at our office in Kirkland, Washington.
- Design, apply and integrate analytical strategies to identify fraudulent activity of all kinds.
- Data analytic responsibilities include pattern discovery, outlier detection, sample design, identification of appropriate analytic and statistical methodology, model development and documentation of process and results.
- Develop working prototypes of algorithms.
- Evaluate the performance of various algorithms, models and strategies based on the real-world data sets.
- Oversee implementation of fraud detection and predictive models in production.
- Degree in applied math, statistics, machine learning or computer science. Ph.D. is preferred but not required.
- Fraud modeling experience required.
- Passion for solving challenging analytical problems.
- Ability to quickly qualitatively and quantitatively assess a problem.
- Ability to work productively with team members, identify and resolve tough issues in a collaborative and respectful manner.
- Experience applying machine learning techniques to real-world problems.
- Fluency in at least one scripting language such as Python, R and Matlab.
- 2-3 years of professional experience is preferred but not required.
- Experience with Hadoop, HDFS, Spark or HBase is preferred but not required.
- Experience with large data set processing is preferred.