Twitter is seeking a motivated and skilled individual to analyze and build fraud detection systems into Twitter's advertising platform. You should have a strong background in statistics and modeling, advertising fraud, and working with large datasets. Every impression and engagement with Twitter's Promoted Products will pass through the system you build, so you should love using data to fight the fraudsters.
Responsibilities
Build complex fraud detection models that scale to run over terabyte-sized datasets
Write and interpret complex SQL queries for standard as well as ad hoc data mining purposes
Write and interpret map-reduce style data analyses using Hadoop and Pig
Work closely and iterate quickly with the revenue team, the analytics team, and other product teams
Code using a mix of SQL, Pig, scripting languages, and stats packages
Summarize and report key analytical findings in both oral and written form
Work with large unstructured and structured data sets. Large, multi-terabyte, billion+ daily transaction volumes.
Requirements
M.S. in Computer Science, Mathematics, Statistics, or equivalent experience.
3+ years of data analysis experience
Experience with large datasets and map-reduce architectures like Hadoop
Experience with scripting languages, regular expressions, etc.
Interest in discrete math, statistics, and probability
Experience in mapping business needs to engineering systems
Follow twitter.com/JoinTheFlock for more recruiting info.
Twitter - 16 months ago
- save job
-
block
