Experienced software engineer with focus on data science problems at the interface of math/stats and computer science. Architected and implemented petabyte-scale big-data software. Experience with machine-trained statistical modeling.
Personal traits: creative, fast learner, passionate.
Worked on big data software on Google's search indexing pipeline.
Quantcast is the hedge-fund of online advertising markets. Promoted to "Quant" role building statistical models. Tasked with developing Quantcast's 2nd-generation technology integrating real-time signals into our algos for high-frequency pricing updates.
Joined as the second member of Quantcast's data pipeline team. I was responsible for re-architecting our core Hadoop MapReduce pipeline to achieve scalability and production-worthiness.
TA'd undergrad and master's courses in algorithms, Java, and C++. Nominated for teaching prize.
B.A. (Honors), Mathematics; University of Chicago, 2009
Each summer, awarded stipend from department to fund undergraduate mathematics research and take graduate courses.
In addition to theoretical and applied mathematics, took coursework in analysis of algorithms, distributed computing, machine learning, probability theory and statistics.
1600 SAT score
Very strong programmer. Developed large projects in Java, C, and C++. Shell and Python scripting. R. Haskell and Clojure. SQL.
Distributed programming: wrote petabyte-scale Hadoop MapReduce jobs at Quantcast. Experience with MPI at college.
Very strong *NIX credentials. I'm a shell CLI guy; I'm handy with sed/grep, paste/join, xargs. I write a lot of code in emacs.
Git evangelist and hand-holder within Quantcast.
Techniques I've used include: