ANNA XIA

Phone: 412-***-**** Email: aczb82@r.postjobfree.com

**** *. ******* ** ***** Highway, Arboretum Plaza One, Austin TX 78759 EDUCATION

Carnegie Mellon University, Pittsburgh, PA Sep 2014 - Dec 2015 Master of Computational Data Science (formerly: Very Large Information Systems), SCS Beijing U of Posts & Telecommunications(BUPT),Beijing, P.R.China Sep 2010 - Jul 2014 Bachelor in Computer Science & Engineering, GPA: Top 5% TECHNICAL STRENGTHS

Programming Languages Python, Java, C, C++, R, Golang, Shell, SQL, Scala, JavaScript Platform & Tools Experiences Hadoop, Spark, Amazon Web Services, PostgreSQL, Thrift, Matlab, Weka PROFESSIONAL EXPERIENCE

WorldQuant LLC, Austin, TX Mar 2016 - now

Core Distributed Service: Parallel Computation of Large-scale Correlation Check Software Engineer

Responsible for leading new feature development, release, test, support and maintenance.

Read code base of 10k lines, refactored 5k lines, Adding cache to reduce correlation server restart time by 80%

Improved service

exibility, response time via pro ling, performed data analysis and visualization, parameter tuning, enhanced reliability and robustness via better monitoring and more automated troubleshooting. Uber Technologies, San Francisco, CA May 2015 - Aug 2015 Automated Document Censor, Machine-learning based classi cation Software Engineer Intern

Designed and constructed document classi er, achieved o ine training, model persistence, online prediction.

Achieved 95% accuracy by speci c data preprocessing and feature engineering, provided great user-friendly access.

Reduced per-document-check-time from 5 minutes to 10 seconds, with censoring accuracy improved by 37%. Tsinghua University, Beijing, P.R.China Fall 2013 - Fall 2014 Boosting MapReduce Performance on Heterogeneous Cloud System Researcher & Developer, part-time

Carried out experiments on Amazon EC2 and local testbed TsinghuaCloud, analyzed data trend and pinpointed causes, demonstrated improvement of job completion time, by 40% in comparison with Hadoop native scheduler, 8.9% with the H-MARES (in INFOCOM).

SELECTED PROJECTS

Auto-Provisioning For Spark Applications On AWS Big Data Systems Studio(15648), CMU,Spring 2015

Pro led representative applications into classes, proposed metric sets for provisioning Spark applications.

Performed load and scale testing and evaluation, provided visualized recommendation of resources provision schema (platform con guration parameters) to meet users’ performance expectations and budget constraints.

Greatly facilitate non-expert users, demonstrated this Cluster Resources Provisioning Recommend System can achieve up to 15X speedup at similar cost, less than 10% prediction error in best performancecost ratio. Hybrid Cloud File System - with Deduplication, Cache, Snapshot Storage System(15746) Fall 2015

Designed and implemented a hybrid Fuse-based le system, leveraging both SSDs and Cloud Storage.

Achieved block-level Deduplication on cloud, added Cloud File System cache, provided Snapshot for recovery. Graph Mining on PostgreSQL Multimedia Databases & Data Mining(15826), CMU, Fall 2014

Accomplished graph mining tasks (PageRank, K-Core, Triangle Counting etc) on real social network graphs.

Achieved high e ciency by applying approximation algorithm, using python and PostgreSQL built-in language. KDD99 Challenge, using R, RHadoop Big Data Systems In Practice(11675), CMU, Spring 2015

Designed and implemented Network Intrusion Detector Learner using R with typical classi er models.

Accelerated intrusion detection e ciency by 3X through RHadoop. Improved prediction accuracy to over 90%.

Contact this candidate