Engineering Assistant

GIRISH VAIDYANATHAN NATARAJAN

**** **** ****** ** *** 1234 Folsom, CA – 95630 210-***-**** *******************@*****.*** www.linkedin.com/in/girishvaidyanathann

EDUCATION

University of Texas at San Antonio MAY 2018

Master of Science (M.S), Electrical and Computer Engineering 3.91/4.0 ANNA University JUNE 2015

Bachelor of Engineering (B.E.), Electronics and Communication Engineering 3.5/4.0 SKILLS

Languages: C, C++ (Proficient), Python, SQL, HTML, CSS, JavaScript, Shell Script (Novice) APIs/Compiler Directives: REST, MPI, Standard Template Library (STL) POSIX threads Software Tools: Microsoft Visual Studio, MSSQL Server, Version Control (Git),Docker,Puppet Courses: Engineering Programming Data Structures and Algorithms Cloud Computing Computer Architecture Operating Systems WORK EXPERIENCE

Graduate Research Assistant-University of Texas at San Antonio 2016-Present

● Integration of Health tracking applications and E-learning environments for Cloud Based Promotions

● Designed and developed a novel and highly accurate data flow protocol for integrating health monitoring applications and e-learning environments like Fitbit, MyFitnessPal and Moodle. The collected information is retrieved in a centralized cloud dedicated for health promotions; providing feedback using dedicated messaging system.

● The data from different applications is extracted using Rest API, preprocessed using panda’s data frame and uploaded to the cloud based dashboard.

● The implemented system and servicing protocol minimizes personnel overhead of large-scale health promotion campaigns and is scalable to assist automated interventions, from automated data retrieval to automated messaging feedback. Graduate Teaching Assistant for Engineering Analysis - University of Texas at San Antonio May 2016 – Dec 2016 PROJECTS

Cache Performance Simulator using C++

● Developed a simulator that studies the behavior of cache based upon the real time memory traces taken from x86 processor.

● Devised the simulator by computing the cache configuration parameters like cache size, block size, associativity and a priority queue implemented using Least Recently Used (LRU) policy. Optimized the cache performance by achieving a hit rate of 85% when benchmarked with different cache configuration parameters. Implementation of JOS Exo-Kernel Operating Systems on x86 Platform

Writing parts of this OS for Intel x86 architecture in C, which implement: booting, memory management (using a 2 level page table), user-level environment and preemptive multitasking. Involves use of GDB for debugging. MPI programming with Python on Linux Platform

● Implemented parallel programming techniques with message Passing Interface (MPI) in Python to parallelize the program that calculated the total number of stars in high pixel; high density image of Andromeda galaxy.

● Optimized the number of CPUs to achieve better execution time in terms of runtime of multicore processors. MIPS Disassembler using C++

● Designed a MIPS disassembler that converts the MIPS binary instructions to machine level instructions.

● The input configuration file consists of binary executable instructions which are analyzed by a middleware program and written to a file as corresponding machine level instructions. Priority Readers and Writers on a UNIX Platform:

Developed a multithreaded C program that gives priority readers priority over writers concerning a global shared variable. The program is capable of supporting multiple readers and writers using mutexes to synchronize access to shared variable. Library Management System in C++

● Developed an automated library system which performs library procedures that provides student and administrator capability.

● The console application made use of object oriented concepts and capable of performing extensive operations like searching a book by ISBN or author, list of available books, adding or removing a book or student from the database. Facial key point detection using Tensor flow

● Employed the existing state of the art Lenet-5 model to detect the facial key points for face images using Google’s open source platform for data programming (Tensor flow).

● Modified the existing models by training a set of 7000 gray scale images; changing the number of convolutional and hidden layers. Performed analysis to find the effective number of these layers which increases quantitative performance.

● Achieved an accuracy of 85% when tested with 4000 new test images which had a significant improvement compared to the state of the art mode which had many false positives for multiple faces on gray scale images. Twitter Sentiment Analysis using Python

● Developed an algorithm to analyze, identify and classify the sentiments of the tweets from GOP presidential election database taken from Kaggle (an online resource addressing machine learning problems).

● The algorithm parses the tweets from the database, makes use of NLTK frequency distribution for extraction of features and trains the model using Naïve Bayes Classifier.

● The model was validated for various test cases which had an accuracy of 93% in predicting the negative words and 64% on positive words which mostly occurred when the tweets had an ironic or sarcastic reference. Blackjack Game and Tic Tac Toe Game using Python

● Programmed a UI friendly tic tac toe game board against a computer program (Artificial Intelligence) that can intelligently respond to players move.

● Simulated a text based blackjack game using top down functional design and implemented all the necessary functionality and features for the board game.

Contact this candidate