Data Engineer

Padma Girish

********@********.*** 240-***-**** Portfolio - https://pgirish.github.io

Technical Skills

Programming: Python, R, PHP, Javascript, HTML, SQL (Oracle, MySQL) Tools & Frameworks: Scikit, Numpy, Pandas, Spark, Spark Streaming, R-Studio, Fusion Charts, Zeppelin Data Science: Data Analysis, Predictive Modeling, Data Visualization Certifications: Data Science Specialization (Johns Hopkins at Coursera), Machine Learning (Stanford at Coursera) Significant Experience

Hughes Played Collected Generated Developed Network a critical data notifications visualization periodically role Systems, in developing when using Software from thresholds Fusion Carrier data Engineer Charts analysis Signals, are breached and and ISPs, XML reporting Earth or for data monitoring Stations is applications missing and and based Access usage for monitoring on Gateways data analysis a in of Satellite PHP collected 2008 and Network data SQL – 2012 Social Responsible Designed Developed 6D, Senior the the for Database User developing Software Interface Schema Engineer a LAMP using and Javascript, implemented based architecture HTML, the Data CSS using and Access MVC backend Layer paradigm processing for MySQL for a Social with based PHP Network database 2012 – 2017 Data Analysis and Machine Learning Project Portfolio

Twitter Loaded Performed Collected Data real-Twitter aggregation Analysis time data data using into and on Spark real-set Spark up time Streaming live Streaming tweets TCP port by and a performed to number feed data of distributed meta-though data as filters analysis a continuous such using as device stream Structured type, for analysis Streaming hash-tags, API etc

NYPD Loaded Wrote Performed Vehicle distributed massive data Collision visualization dataset analyses Data on to New using understand Analysis York Zeppelin City using trends vehicle to Spark more in collisions factors clearly such into demonstrate as Spark fatalities, Dataframe insights locations from to perform for large-over scale analysis 1M collisions analysis Forecasting Applied Evaluated Performed ridge Housing the data performance regression analysis Prices and for of through an visualization linear optimal and Advanced model non-of home linear of prediction Regression sale regression prices and Techniques using models achieved data to predict an frames accuracy home with of Pandas sale 0.84% prices and r-squared using Matplotlib Kaggle SciKit value Predicting Predicted Compared Utilized Random interest interest various Forest level level algorithms of of classifier new new including Rental Rental to train Listings Listings Bagged the model data Decision from and achieved RentHop Trees, Random an using accuracy Ensemble Forest, of 0.AdaBoost, 725% Learning Gradient Boost Kaggle Predicting Implemented Employed Developed Next-an a Shiny n-an word order n-grammodel based Markov using UI Natural to model accept to predict which Language input the uses words most Bayesian Processing from likely the next inference user (word NLP) and via using to the display NLP Katz’s techniques predicted back-o next methodology word Forecasting Implemented Developed Post a model a multi-Foreclosure to variate forecast linear Housing the the regression market Prices that rate considered for homes undergone pre-foreclosure foreclosure factors to from predict FannieMae house sale data price Education

Master of Science in Information Technology - University of Maryland University College (UMUC)