Created pipeline to process hundreds of terabytes of sessions into usable data for A/B testing. Created numerous alerting filters to detect all types of anomalies within YouTube.

Making numerous analytics on different types of data sets. Regularly using Spark, Keras, TensorFlow, NumPy, Pandas, etc..Teaching classes in Hadoop, Spark, and Data Science.

- Implemented Metropolis-Hastings Algorithm to create probabilistic model of a classified network.- Created functional model to determine lethality descrimination of missiles under contract with MDA.

Building analytics which find anomalies in network logs, designed to find malware, recognize attacks, and enable alerting. Part of deployment effort to stand up analytics on DOD systems.Created training manual and CD for specific network analysis tool (Noesis).

• Built analytics which present massive data sets to clients and customers in a simplified manner that saves time and advances spatial relationships and data relevance. Uses mainly Hadoop MapReduce technology and Cloud Base. • Used advanced methods of data mining to surface sets of data and creates the tools, which then replicate those actions. • Established streaming sets of data delivered to corporate servers for the use of analytics and research and development. • Continued to develop skills in Java with on-the-job training and performing programming tasks.

• Worked closely and successfully with computer programmers, engineers, analysts, and other mathematicians. • Evaluated several years of scientometric data to find new trends in technology research. This involved detailed analysis of giant data set using several types of clustering algorithms. Managed database, computer science interns, and full-time employees as part of this project. Presented award for outstanding management of employees. • Used state-of-the-art graph algorithms to investigate structure of specific classified contact networks. Found structure in networks which, in turn, found many people of interest that were previously unnoticed within analysis. • Invented algorithm to sift through massive amount of a specific type of data which resulted in funding for brand new question-focused database, and creation of several new government careers. Subsequently co-led entire research group dedicated to finding more points of interest in the new data. • Presented with monetary and time-off awards for tremendous value of findings, quote on award said “greatly exceeded what was believed to be possible.” • Completed Cryptologic Mathematics Program in 2009. Presented modification of classified algorithm at several conferences. Presented at three SIGDEV conferences 2008-2010.

Doctoral student in Algorithms, Combinatorics, and Optimization.