Assistant Engineer

SAM SHAH

King St., San Francisco, California

Mobile: - Email: *******@*****.***

http://www.eecs.umich.edu/ shahsam

E University of Michigan Ann Arbor, Michigan

Ph.D., Computer Science (So ware Systems) May

Dissertation Title: Leveraging Context for File Search & Organization

Committee: Brian Noble (chair), Atul Prakash, Mark Ackerman, Lada Adamic

M.S.E., Computer Engineering April

B.S.E. (Magna Cum Laude), Computer Engineering April

Honors: Department of Education GAANN Fellowship, Dean s List, University Honors.

E Platform Engineer (Analytics) Aug. Present

Ning, Inc. Palo Alto, California

Clickstream Analysis

Built a tool that sessionizes and prunes tra c to see how users move throughout the site.

Responsible for identifying several broken ows, resulting in double-digit tra c growth

Engagement, Growth, and Other Business Metrics

Helped develop advertising dashboard, which classi es ad inventory by content catego-

rization and demographic features

Developed self-serve options for engineers and executives to access data and metrics (e.g.,

tra c by geo-IP location, network creations, etc.)

Network Recommendation

Built network discovery tool that allows users to nd social networks via keyword search

Category generation for Ning s network recommender

Topic Trends & Sentiment Analysis

Created a zeitgeist tool that mines blog posts and comments to determine trending topics

and user attitudes.

Multivariate & Usability Testing

Lead on developing an A-B/MVT testing framework for developers and QA

Scaling these tools and work ows across a TB dataset using Hadoop/MapReduce

Ad-hoc analysis, mainly using Pig (including custom UDFs), streaming, and/or the Java

MR framework, coupled with R

Collaborating with product and engineering to determine new metrics and analyses

Interfacing with executives to understand growth and usage patterns across the platform

Created infrastructure for importing (ETL) and managing data in HDFS

Operation and maintenance of a -node Hadoop and -node HBase cluster

Research Assistant Sept. May

University of Michigan Ann Arbor, Michigan

Context Sharing

Designed and built a decentralized architecture for stitching context together across a

user s social neighborhood, aiding spontaneous ad-hoc collaboration.

Developed a new abstraction for specifying personal boundaries on disclosure.

Performed a user study, which suggests the system is palatable to users, and performance

analysis, which indicates the additional overhead is negligible.

SAM SHAH

King St., San Francisco, California

Mobile: - Email: *******@*****.***

http://www.eecs.umich.edu/ shahsam

Context-Enhanced Desktop Search

Designed and implemented a system that captures context from the user to e ectively

reorder and extend search results from personal desktop search tools.

Conducted a eld study with this prototype implementation, which found that users nd

improvements in their search results with the system.

Email Patterns

Performed a large-scale study of email, including message payloads.

Teaching Assistant Jan. Apr.

Taught advanced C++ concepts (e.g., advanced templating, STL, etc.) and OOP patterns.

Responsibilities: Code reviews for student projects; held o ce hours; graded exams and

homework. ( students)

Teaching Assistant Sept. Dec.

Taught an introduction to data structures as well as how to think like a computer scientist.

Responsibilities: Lead discussion sections; held o ce hours; graded projects and exams.

( students)

S Large Data

Hadoop/MapReduce, Pig, Streaming, Java MR framework, C++ pipes.

Distributed Systems

Scalable systems, eventually consistent systems, DHTs, gossip protocols.

Languages

C, C++, Python, R, Java, SQL, XML/RDF, Bash, LTEX, some LISP.

A

Development Skills

Big data, le systems, distributed systems, kernel, threaded, web and semantic technologies.

Statistics & HCI

Strong knowledge of statistics and usability analysis (user studies).

P S. Shah, C. A. N. Soules, T.J. Giuli, and B. D. Noble. Ligature: An Architecture for Sharing Context.

Under submission.

S. Shah, C. A. N. Soules, G. R. Ganger, and B. D. Noble. Using Provenance to Aid in Personal File

Search. In USENIX, pages, Santa Clara, CA, USA, June .

S. Shah and B. D. Noble. A Study of Email Patterns. So ware: Practice and Experience, :,

.

Contact this candidate