Resume

Sign in

Data Project

Location:
Dallas, Texas, United States
Salary:
70000
Posted:
February 21, 2018

Contact this candidate

Nitesh

Srivatsav

Aspiring Data Scientist

Machine Learning enthusiast

Personal Info

Address

**** ********* ****,

Dallas,TX-75252

Phone

214-***-****

E-mail

ac4j5e@r.postjobfree.com

Github

https://github.com/TheCode2017

LinkedIn

https://www.linkedin.com/in/nitesh-srivatsav/

Tools Skills

Java,Python(Scikitlearn,Pandas,Numpy,NLTK),

Scala

Advanced

Git,SQL,MySQL,NoSQL

Proficient

Spark(Mllib,Pyspark),Kafka,Hadoop

Proficient

Mxnet,Caffe,TensorFlow

Proficient

Pytorch

Proficient

Elasticsearch-Kibana,Matplotlib,seaborn

Advanced

ML libraries(DLIB,SHOGUN,NLTK)

Proficient

Amazon AWS EC2

Proficient

HTML5,Javascript,CSS3,D3,Plot.ly.

Proficient

Linux,Mac,Windows

Advanced

Hard Skills

Data Analysis

Statistics

Data visualization

Machine learning(SVM,Naive Bayes etc)

Image processing

Certificates

2017-06

Machine Learning:Hands-on in Python and R In

Data Science-Udemy

2017-07

Getting and cleaning data-Coursera

2017-07

Exploratory data analysis-Coursera

Junior data scientist and a Masters student in data science with 1+ year of experience in project work and freelance jobs. Developed deep learning models for object classification and detection and also skilled in machine learning, statistics, problem solving, and programming.Looking for a Data Scientist role in a company where I can contribute and also improve upon my skills.

Experience & Projects

2017-11 -

present

Freelance data scientist

• Worked on freelance data jobs on Upwork.

• Built various models(random forest,SVM) for different jobs. Gained immense experience on cleaning dirty datasets and finding the best classification/regression models to fit on them.

2015-05 -

2015-07

Paragon Digital Services

Worked on a project to develop contextual and targeted keywords to generate efficient ad-words for various products.

Participated in requirements meetings and data mapping sessions to understand business needs and to build ad-words based on the information obtained from the sessions.

Scraped advertisement data from the web(from search results) using BeautifulSoup and analyzed the different ad-words used for various products.

2014-05 -

2014-07

HCL technologies

Completed a sentiment analysis mini-project that consisted of collecting and analyzing reviews and produced an overall review of the product.(Using asp.net and MS-SQL).

Collected numerous reviews of a restaurant from the web and stored them in the MySQL database.Then using asp.net, built a tool that removed the stop words and using MeaningCloud determined the sentiment of each review and displayed the total number of positive and negative reviews for the restaurant.

2017-02 -

2017-05

Big Data Project “Crime rate Forecasting System"

Developed an unsupervised machine learning technique with an external knowledge base as our input to the Forecasting system.

Used the Spark MLLIB(K-means clustering) for clustering the dataset to create a structured input for our forecasting system.

Used the ARIMA model(Autoregressive Integrated Moving Average) that gave me an accuracy of 72% when the forecasted values were compared to the original data.[Scala,PySpark]

2017-04 -

2017

Big Data Project "Twitter Sentiment Analysis"

• Used a scraper to live stream data from Twitter and filtered the tweets on certain keywords. Then using Apache Kafka performed sentiment analysis on the live stream of Twitter data using StanfordNLP library to classify the tweets as 'positive','negative' or 'neutral' and also plotted the number of tweets of each category using Elasticsearch's Kibana.[Pyspark]

2017-09 -

2017-12

Object detection using TensorFlow

• Used the TensorFlow API to successfully detect objects in an image/video. Trained new objects on the API(soccer ball) using Amazon AWS EC2 for faster training(reduced training time by 20%).

Achieved a remarkable 94% accuracy on the test images and finally used this trained model to detect the objects in a video game ( GTA V, Watchdogs). [Python,Tensorflow]

2017-09 -

2017-12

Object classification using Pytorch

Used the Pytorch library build a CNN classifier(2 convolutional +2 pooling+1 fully connected layer) to classify the Yelp restaurant dataset(200,000 images=10GB).

Used Amazon EC2 (to reduce training time)trained about 20,000 images(due to limited computational power) on the CNN classifier and achieved a 75% accuracy on the test set.Would have achieved better accuracy if I had better computational power.[Python]

2018-01 -

2018-01

Analysis of the 2017 soccer transfer market

A pet project where I scraped data from the web using BeautifulSoup and MS Excel and analyzed the number of transfers made by each club.

Also ran complex Spark SQL queries on the scraped data to extract out useful information on the transfers made by teams in the market.[Python]

2017-07 -

2017-07

Exploratory data analysis

Analyzed the National Emissions Inventory (NEI) dataset to discover the effect of PM2.5(air pollutant) over the years in the United States through the use of various plotting systems in R(base, lattice, ggplot2). [R]

2017-07 -

2017-07

Coursera project on cleaning data

Given untidy Samsung S2 datasets, cleaned the dataset of Null values and reshaped the dataset(using reshape) and successfully derived useful insights from the data. [R]

Education

2016-08 -

2018-05

The University of Texas at Dallas,Masters in Computer Science,Data Science

• Excelled in machine learning and data science coursework. Coursework: Machine learning( CS 6375),Big data management(CS 6350),Statistical methods for data science(CS 6313),Database Design(CS 6360),Algorithm Analysis and Design(CS 6363),Data Analysis.

• GPA: 3.62/4

2012-08 -

2016-05

SRM University,Computer Science

GPA:3.6/4

HONORS AND AWARDS: President of the ML club,Dean's Excellence Scholarship,Top 5% of the class.



Contact this candidate