Mike Perkowitz

6910 Roosevelt Way NE, #124

Seattle, WA 98115

206.915.3138

mike@perkowitz.net

Summary

I am seeking an opportunity to use my technical and business skills to design and build innovative applications for real users, particularly in the areas of search, social networking, and information management. I have extensive experience with technical innovation, particularly in data mining and internet applications, as well as in anticipating and designing for user needs, and am skilled at bringing together technical and business considerations to create a product vision and plan.

Technologies        

SQL/MySQL, PHP, Perl, Javascript, Java, LAMP.

Professional Experience

Program Manager, Trailfire, Inc.                                                  November, 2005 - Present

Accomplishments:

§       Worked on product strategy and created feature specifications for our web-based social annotation tool.

§       Participated in crafting overall company strategy, planning and executing marketing initiatives, and evaluating potential investors and partners.

§       Designed and implemented web-based user interface.

§       Hired and worked with graphic designers on product look and feel.

§       Developed clustering and other data mining algorithms for analyzing and organizing web pages.

Software Design Engineer, Amazon.com                                June 2004 – November 2005

Accomplishments:

§       Designed and implemented a personalization engine for tailoring advertising for Amazon affiliate web sites to site content and traffic.

§       Explored new tools and algorithms for understanding customer behavior to better recommend relevant products.

§       Maintained a detailed customer behavior database, created tools for viewing and analysis of that customer behavior, and assisted others throughout the company in better understanding our customers.


Researcher, Intel Research Seattle                                          November 2002 – June 2004

Accomplishments:

§       Helped design a system for recognizing and identifying everyday human activities among residents of assisted-living facilities.

§       Developed algorithms for extracting structured statistical models of everyday human activities from unstructured web sources for use in activity recognition, and created a partial knowledge base of activities.

§       Explored possible uses of mobile technology to bring people together based on online relationships.

Senior Engineer, Advanced Development, Appliant, Inc.           August 2000 - July 2002

Accomplishments:

§       Prototyped new features and reports and wrote specifications for implementation.

§       Created automated diagnostic tools for pinpointing problems in web site infrastructure.

§       Designed data analysis and reports showing the connection between web site performance and customer retention and sales.

§       Performed in-depth analysis of customer data and prepared explanatory presentations.

§       Co-wrote white papers and a conference paper explaining some of Appliant's technological innovations.

Research Assistant, University of Washington Department of Computer Science

                                                                                                   September 1993 - August 2000

Accomplishments:

§       Developed new statistical and clustering algorithms for analyzing user behavior data (Ph.D. dissertation work).

§       Applied these algorithms in a system to analyze user traffic at a web site and adapt the site structure to facilitate navigation.

§       Developed an algorithm for extracting structured information from semi-structured internet information sources, and implemented a prototype system (M.S. work).

§       Published numerous papers in artificial intelligence, machine learning, and data mining journals and conferences, and presented results at conferences.

Research Intern, Microsoft Corporation                                   June 1996 - September 1996

Accomplishments:

§       Developed working prototype of a system for sharing web-based information via a character-based user interface.

§       Published and presented a paper describing this work at the 1996 Lifelike Computer characters conference.

§       Received a patent for this work.


Education

University of Washington, Ph.D., Computer Science, February, 2001.

University of Washington, M.S., Computer Science, June, 1995.

Brown University, B.S., Cognitive Science, May, 1993.

Selected Publications

M. Philipose, K. Fishkin, M. Perkowitz, D. J. Patterson, D, Fox, H. Kautz, and D. Hahnel. Inferring Activities from Interactions with Objects. IEEE Pervasive Computing. Vol. 3, num. 2, 2004. pp. 50-57.

M. Perkowitz, M. Philipose, K. Fishkin, and D. J. Patterson. Mining Models of Human Activities from the Web. Proceedings of the Thirteenth International World Wide Web Conference (WWW 2004).

J. B. Chen and M. Perkowitz. Using End-User Latency to Manage Internet Infrastructure. Proceedings of the 2nd Workshop on Industrial Experiences with Systems Software, 2002.

M. Perkowitz. Adaptive Web Sites: Cluster Mining and Conceptual Clustering for Index Page Synthesis. Ph.D. dissertation, University of Washington, 2001.

M. Perkowitz, and O. Etzioni. Towards Adaptive Web Sites: Conceptual Framework and Case Study. Artificial Intelligence Journal, 1999.

M. Perkowitz, and O. Etzioni. Adaptive Web Sites: Concept and Case Study. Communications of the ACM, 1999.

M. Perkowitz and O. Etzioni. Adaptive Web Sites: Automatically Synthesizing Web Pages. Proceedings of AAAI 1998.

M. Perkowitz, R. Doorenbos, O. Etzioni, and D. Weld. Learning to Understand Information on the Internet: an Example-Based Approach. Journal of Intelligent Information Systems. Vol. 8, num. 2, 1997. pp. 133-153.

E. Charniak, C. Hendrickson, N. Jacobson, and M. Perkowitz. Equations for Part-of-Speech Tagging. Proceedings of AAAI 1993. pp. 784-789.

References

Available upon request.