Doc - Carnegie Mellon University

Document Sample
Doc - Carnegie Mellon University Powered By Docstoc
					                                                                                                               Phone: (+1) 412-818-6804
Mayuresh Vartak                                                                                        Email: mvartak@andrew.cmu.edu
                                                                                            http://www.contrib.andrew.cmu.edu/~mvartak

OBJECTIVE:
To obtain full time position in software design, development and research; where I can utilize my technical and analytical abilities.
EDUCATION:
Carnegie Mellon University, Pittsburgh, PA                                                                                   May 2010
Master of Science, Information Networking
Visvesvaraya National Institute of Technology, Nagpur, India                                                                 May 2007
Bachelor of Technology, ECE
PROFESSIONAL EXPERIENCE:
Graduate Research Assistant, Institute of Complex Engineered Systems, Carnegie Mellon University                Dec 2009- present
 Design and Development of Health-Node, an intelligent communal kiosk integrated with sensor system (C# and MySQL).
Graduate Research Assistant, Machine Learning Department, Carnegie Mellon University                            Jan 2010 - present
 Exploratory analysis of Twitter data to extract text usage patterns, user relations using natural language processing techniques
    such as Vector Space search. Clustering and leaning models (Lucene, Carrot2, LingPipe, Mahout).
Software Developer Intern, Bloomberg LP, New York                                                             Jun 2009- Aug 2009
 Designed and developed automation tools for ticker scanning and Bloomberg identifier mapping in C++ and Perl.
System Engineer, Tata Consultancy Services Limited, Mumbai India                                              Jun 2007- July 2008
 Served as module lead in General Electric Capital Solutions, Europe and was responsible for design and implementation of
    automation and enhancements in core business application for its entire development cycle (C, C++, PL/SQL).
TECHNICAL SKILLS:
Language: C, C++, Java, C#, Assembly X86, MATLAB, Shell scripting, VHDL, Perl, Python, JSP, Servlets, PL/SQL
Platforms & Tools: UNIX, Windows, Eclipse, Oracle Designer, gdb, Visual Studio
RELEVANT COURSES:
Operating Systems, Distributed Computing Systems, Advanced Storage Systems, Machine Learning, Search Engines & Web Mining,
Computer Vision, Web Application Development, Algorithm Design & Analysis, Database Management. Computer Networks
ACADEMIC PROJECTS:
FSCK Utility (2010)                                                                                                                 C
Created a rudimentary fsck tool to identify, parse, read, and manipulate the on-disk image of an ext2 file system.
Kernel, Kernel Debugger and Thread library (2009):                                                                    C, X86, Simics
Designed multithreaded preemptive kernel based on X86 architecture, which supports basic system calls, virtual memory and
keyboard, console, timer device drivers implementation. In-Kernel debugger is developed with functionality to inspect and manipulate
kernel state. Kernel supported thread library is implemented to support threads and locking.
Distributed File System / Flash and Disk File System (2010):                                                                     C++
Implemented distributed file system based on ‘Frangipani’ architecture, consisting of lock server for consistency, extent server for
files storage supporting basic file system operations using FUSE and RPC library.
Distributed Password Cracker (2009):                                                                                             C++
Designed and developed a distributed password cracker based on UDP, where the server handles and assigns work unit to multiple
clients. Error checking, fault tolerance and synchronization have also been implemented to make the system robust to failures.
Hadoop (2009):                                                                                                                   Java
Analyzed Twitter hash-tag and user data on Hadoop clusters using Map-Reduce programs.
Search Engines (2009):                                                                                                           Java
Implemented indexer, and search models such as Boolean, Vector Space, Query Likelihood (probabilistic), Global Page Rank (GPR),
Personalized Topic Sensitive Page Rank (PTSPR). Their performance is analyzed and compared using Trec-Eval.
Cognitive State Classification with Magnetoencephalography (MEG) Data (2009)                                                   Matlab
Built single subject classifier to learn word categories from the MEG readings for given set of words based on different models such
as Gaussian Naive Bayes with PCA, Logistic regression and Support Vector Machines (SVM) and analyzed their performance.
Object Segmentation and Recognition (2009):                                                                                    Matlab
Designed and implemented apples counting technique from set of images, using L*a*b color space thresholds (hysterises),
morphology and mean shift algorithm. It achieved 0.82 F1 measure and was selected in Hall of Fame for its performance.
Not So Idle Scan (2008):                                                                                               C, Shell Script
Developed idle scan technique based on normal distribution, for zombie hosts with little external traffic.
HONORS:
   Recipient of academic scholarship covering $23,000 tuition fees from Information Networking Institute, CMU.
   Received National Talent Search (NTS) Scholarship awarded by NCERT to top 1000 students across India in 2001.
   Qualified Regional Math Olympiad (RMO) and was among top 30 from Mumbai in 2001.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:10
posted:9/21/2011
language:English
pages:1