Results-oriented technical director with extensive software-engineering experience. Broad range of knowledge and skills, including data warehousing, data visualization, computer graphics, data mining, machine learning, web application design and development, formal product specifications and documentation, and project life-cycle management. Programming experience in Python, Java, Perl, C/C++, PHP; extensive experience with the LAMP architecture; authored custom apache modules; extensive experience with raw HTML/XHTML/XML/CSS; experience with many database solutions, including MySQL, Oracle, Netezza, Sybase, and Berkeley DB. Familiar with MATLAB and Quadstone tools.

Head of Technology Recommender Systems @ We designed and developed a large-scale recommender system for performers, events, and tickets. Our modular design allows us to combine scores from behavioral and content-based models. Unlike a static catalog of books, movies or retail merchandise, events take place at some location at a particular time, and so event recommendation posed interesting geographical and spatial information retrieval problems for us to solve. From May 2013 to Present (2 years 8 months) Greater Boston AreaSenior Director, Personalization and Targeting @ • Implemented end-to-end online multivariate experiment framework for CBSi sites including statistical evaluation, reporting, and user bucketing web service. • Ran hundreds of experiments generating millions of dollars of incremental annual revenue. • Gave several invited talks on multivariate testing at industry conferences. • Deployed revenue-optimizing web-service for cost-per-click offers that generated a sustained 28% increase in revenue per offer compared to the initial display strategy. • Oversaw the development of a recommendation service using ensemble learning to incorporate information from heterogeneous sources (e.g. clickstream, metadata, user-generated content, social, etc.) From April 2011 to May 2013 (2 years 2 months) Director Personalization and Targeting @ Led applied data mining group in creating a recommendation and content optimization API, which handled more 100 million requests per day. From June 2008 to April 2011 (2 years 11 months) Director Personalization and Targeting @ • Pioneered group focused on data exploration and application architecture supporting business unit efforts to increase user engagment and incremental revenue. • Used clustering, association-rule mining, and collaborative filtering to train and test models of user behavior, and developed batch-processing framework and service layer for scoring and delivering data to web applications. • Produced up to a fifteen-fold lift in click-through over control group with user and association-rule based content targeting. From May 2005 to June 2008 (3 years 2 months) Director, Data Warehouse Infrastructure @ • Scaled data-warehouse processing to over four times its original volume, to over 100 million events stored each day. • Managed mission-critical team with diverse objectives, supporting both operational and business-driven projects. • Successfully drove projects to migrate the warehouse from Solaris to Linux, and from Oracle to Netezza, reducing total cost of ownership and increasing processing capacity. • Effectively implemented and coordinated rollout of site-wide P3P privacy policy to reduce impact of the growth of Internet Explorer 6. • Drove several data-mining efforts, providing data for industry leading business intelligence delivered to premier vendors on competitive products, and used web-activity data to predict sales in the retail space. From July 2000 to May 2005 (4 years 11 months) Software Engineer @ • Provided application development in Perl, Java, and C to improve Internet reporting and data mining. • Developed and deployed Web-based workflow clients for intranet and Internet using Oracle backend. • Collaborated with end users to produce custom reports for revenue-based accounts. • Designed application features to integrate with core technology such as TCP/IP network, Web server, and various system management tools. • Used programming expertise to improve enterprise Web server and database architectures; maintained and extended pool of Apache Web servers running custom modules. • Maintained and developed data-extraction applications for clickstream user tracking. • Created and implemented loading/reporting solutions for Oracle data warehouse. From April 1997 to July 2000 (3 years 4 months)

Master of Liberal Arts, Information Technology @ Harvard University From 2000 to 2005 BA, Music @ Wesleyan University From 1982 to 1987 Cliff Lyon is skilled in: Data Mining, Python, Relational Databases, Text Mining, Design of Experiments, Recommender Systems, Landing Page Optimization, Web Analytics, Java, Data Visualization, Predictive Analytics, Statistics, Web Content Optimization, Linux, Data Warehousing