Expanded responsibilities to company-wide product management. Defined product requirements, roadmaps, and marketing for all products, including database discovery, data classification, and threat detection software. Briefed analysts and developed sales collateral, webinars, podcasts, and product demonstrations based on market research and interactions with customers and industry analysts. Executed strategic partner integration with McAfee by navigating their complex partner program. Transformed company sales model from legacy hardware to modular software solution to accelerate sales.

Led the design and implementation of a product for insider threat detection in complex database networks, requiring close collaboration with UI developers, QA, sales, and marketing. Applied behavioral modeling, unsupervised anomaly detection, and categorical clustering to learn high volume transactional environments and identify unusual access to critical data. Led the development of a web-based user interface and drove cutting edge visualization to enable security analysts to easily understand the movement of high value data in their environment. Streamlined machine learning and optimized reporting on billions of SQL transactions within a hardware constrained platform to scale performance to enterprise level customers.

Performed foundational research for behavioral modeling and threat detection in complex database networks.

Lead analytics team in the development of client facing and internal reporting, data warehousing, behavioral modeling, stream-based analytics infrastructure, and segmentation analysis for SaaS e-commerce platform handling $1B+ multi-channel retailers. Worked closely with pre-sales, account management, support engineers, quality assurance, development, operations, and management to grow from a handful of clients to over 150 top e-commerce websites.

Created and patented underlying recommendation algorithms for a real-time personalization and revenue optimization platform. Redesigned algorithm implementation to scale service capacity to 10k+ aggregate page views per second. Designed data warehouse to enable data driven personalization by a shopper’s time of day and geography.

Researched the efficacy of natural language processing techniques for named entity and relation extraction from unstructured text using support vector machines and averaged perceptrons to support U.S. intelligence efforts to collect named entities from foreign texts. Created XPath tools to cull semi-structured data from the web to automate the collection of named entities.

Developed and delivered FalconID application fraud and identity theft models to U.S. and international telecommunication, retail, and banking clients. Designed and implemented Java libraries for flexible, extensible, and fully tested model development. Grew the UCSD data mining contest into an international competition with hundreds of teams by collaborating with UCSD faculty and local industry to create a compelling challenge with real world data.