I write code to solve large scale data analysis an machine learning problems.
Currently I'm working on distributed machine learning system for personalized ad targeting and audience modeling. Building it from the ground-up using best open-source technologies.
I am looking for serious technical challenges in applying machine learning and predictive analytics to structured and unstructured `Big Data`.
• 3+ years developing `Big Data` analytics (Spark, Spark ML, Hadoop, Cassandra, HBase)
• 4+ years functional programming in Scala (Akka, Spray, Finagle, Scalaz)
• 7+ years hands-on Java development (concurrency, io, generics, reflection, JEE)
• statistical computing with R
Senior Software Engineer @ Building machine learning and predictive modeling systems powering Collective audience targeting & optimization at terabyte scale. Using massive data sets to build thousands of models that drive hundreds of thousands personalized advertisement decisions per second. Working closely with Data Science team to “productionize" complex mathematical models and solve technical challenges.
• design and development of next generation distributed machine learning production systems with Spark and Spark ML
• participating in data architecture and engineering decision-making to design a number of ongoing and planned data analytics projects
• evaluating and adapting new approaches and technologies for predictive modeling and machine learning
• building tools for data science team to support ongoing research projects
• applying machine learning to solve big data problems
Technologies: Scala, Spark, Spark MLLib, R, Hadoop, HBase, Python + PySpark From February 2015 to Present (11 months) Greater New York City AreaSenior Software Engineer @ Technical leader in charge of the design and development of Pellucid Analytics data platform. Pellucid is technology startup building tablet- and browser-based software, designed for providing timely, data-driven, expert analysis to investment banking professionals.
• implemented distributed large-scale solutions for financial data acquisition and analysis (fundamentals, pricing, estimates, risk, etc.)
• built point-in-time access to financial data with Datomic and Cassandra
• made possible to ask any question against how the data looked at any time in the past. “What did I know about this company when I made this particular analysis?”
Designed, built, and deployed Pellucid data ingestion and analytics system: Spark, Scala, Cassandra, Datomic, Spray, Amazon EC2. From June 2013 to February 2015 (1 year 9 months) Greater New York City AreaSenior Software Engineer for Deutsche Bank @ As a part of Rates & Credit OTC Derivatives clearing team, I was involved in design and development of next generation OTC derivatives clearing platform. Consolidating data (IM, VM, position, activity, cash, mark-to-market) of OTC products from all CCP Clearing Houses to provide consistent data for all PB reporting teams. Successfully launched new platform to support expansion of the multi-asset clearing across new Clearing Houses around the globe.
• drastically reduced time for building end-of-day reports utilizing Disruptor for new high-throughput trades processing engine From October 2011 to June 2013 (1 year 9 months) Software Engineer for Vehco @ At Kentor I was a part of a team of five developers implementing application for one of leading European transport IT-solutions provider - Vehco. The goal for our team was development of messaging module for Vehco’s product CoDriver which allows fast and reliable message passing between bus/truck drivers and office team. We used agile techniques such as daily standups and weekly iterations in order to provide quick feedback to the business.
• developed heavy loaded server side application working with hundreds and thousands connected mobile clients
• developed rich client web application based on Google Web Toolkit
• mobile application development with Java 1.3 on IBM J9 jvm, and in-house gui framework based on AWT From February 2011 to October 2011 (9 months) Software Engineer @ SdaemSami is local reviews guide on apartments rentals and neighborhoods in Russia (rentals by owner). As a principal software engineer I designed and developed backend application, integration with electronic payment gates. It was build with Spring MVC web framework, Hibernate for ORM, Tomcat as servlet container with Apache HTTPD + AJP as proxy for load balancing and serving static content. From June 2010 to December 2010 (7 months) Software Engineer @ Implemented scalable solutions for various natural language processing problems using Apache Nutch, Apache Hadoop and Weka as primary machine learning library: sentence extraction, part of speech recognition, natural language generation.
System used Nutch for crawling web and collecting unstructured text data, further data mining was done with Hadoop map-reduce jobs and Weka. From February 2008 to June 2010 (2 years 5 months)
Master, Computer Science @ Saint Petersburg State Polytechnical UniversityMaster, Economics @ Saint Petersburg State Polytechnical University Eugene Zhulenev is skilled in: Problem Solving, Software Engineering, Distributed Systems, Machine Learning, Scala, Java, Apache Spark, Hadoop, Big Data, Big Data Analytics, Predictive Modeling, HBase, Cassandra, MapReduce, R
Websites:
http://eugenezhulenev.com,
https://github.com/ezhulenev,
http://stackoverflow.com/users/1177991/eugene-zhulenev