
Software Engineer at Google
Greater Seattle Area

Software Engineer at Google
Greater Seattle Area
I am an engineer with 15+ years of experiences in the information industry, in both engineering and research. I devote myself to building world-class information systems that matter to people. To be more specific, I like to dig into a tremendous amount of data collected all over the world, in order to discover interesting and useful information, like causal relationships, patterns, trends, and so on. These discoveries may change people's minds.
large scale distributed systems, data mining, machine learning, VOIP, telephony
(Public Company; GOOG; Internet industry)
October 2007 — Present (2 years 3 months)
- Google Ad Planner back end
-- AdPlanner is Google's platform to help advertisers planning their Ads campaigns.
- Focuses on estimates of Internet traffic and demographic properties
-- Expertise on large scale data mining, machine learning, map reduce, etc.
- Google Voice's voice mail storage and processing back end
-- Google Voice is a global-scale telephony platform - VOIP, SIP
-- Focused on its voice mail part - a distributed system that plays, records, stores, converts, and transcribes voice mails and call recordings at global scale.
-- Designed for and maintained availability at 99.99% for 15 months in total, before I switch to the next project.
-- Expertise on reliability of replicated services: such as data and service replication, service health check and failing over, load balancing, throttling, production monitoring, capacity planning, etc.
(Educational Institution; 5001-10,000 employees; Higher Education industry)
September 2006 — September 2007 (1 year 1 month)
The two publications below describe what I did as a postdoc fellow at Dartmouth.
- MAP: A scalable monitoring system for dependable 802.11 wireless networks, in IEEE Wireless Communications, pp. 10-18, Oct. 2008.
- Detecting 802.11 MAC Layer Spoofing Using Received Signal Strength, in proc. of IEEE Infocom'08: the 27th Conf. on Computer Communications, Phoneix, AZ, apr 2008
(Higher Education industry)
April 1996 — May 2000 (4 years 2 months)
Lecturer, Dept. of Computer Science;
Researcher, Commtech research institute;
PhD , Engineering Science , 2001 — 2006
During the 5-year pursuing a doctoral degree at Dartmouth, I worked on a new kind of information retrieval system, what we call Process Query System (PQS). "Process" is a high-level abstract of most real-world activities that produce observable time-series. PQS provides methods to query processes from time-series data using statistical models, just like people query database records using an SQL "select" statement.
We successfully prototyped PQS to different application domains:
- network security, for example, intrusion detection systems;
- physical object tracking, consider how radar tracks aircrafts;
- social network analysis, how people connects to each other;
- sensor networks;
- automatic computing.
Through this research project I developed expertises for statistical models and data mining, specifically:
- discrete time series analysis;
- Markov models, Bayesian networks, Kalman filters, and their applications in data mining;
MS , Computer Engineering , 1993 — 1996
B.E. , Computer Engineering , 1988 — 1992