Idea icon

About Me


Hello! I'm Siddharth, a master's student with a high level of expertise in Data Analytics and Machine Learning

I'm pursuing my Masters Degree in Business Analytics at The University of Texas at Dallas and I spend most of my time working on interesting Projects, attending Hackathons or browsing Reddit / Quora!

I am from India and I am currently exploring Dallas, TX


What you should know


This is my education and work experience timeline.

  Download My Résume

You can download my résume in PDF format if you would like.

   Download    View

    GitHub &
LinkedIn

   GitHub    LinkedIn

  Analysis Tools & Programming

Python
Spark
R
Tableau
Hadoop
Power BI
Abinitio
Hive
Excel

  Skills & Others

NLP
Google Cloud
Statistical Modeling
Azure
A/B testing
MySQL
Deep Learning

  M.S. BA Coursework

Applied Machine Learning
Big Data
Programming for Data Science
Applied Econometrics
Predictive Analytics
Prescriptive Analytics
Adv. Business Analytics with R

  • Discover Financial Services

    Lead Data Science Analyst

    • Developed a machine learning algorithm using call transcript, agent notes, customer events, applications, transactions data and successfully categorized 96% of calls into 44 call types with >85-90% accuracy
    • • Re-built monthly forecast model for all deposit lines using the new customer call intent logic resulting in 7% reduction in forecast calls for the next calendar year giving $5mm in operational benefit
    • • Identified a prominent messaging mobile placement and conducted A/B testing using Z-test, thereby increasing visit to messaging rate by 41% and call displacement by 10%

  • The University Of Texas at Dallas

    Masters Degree

    Currently pursuing Masters in Business Analytics at UTD, specializing in Data Science & Analytics.

  • FedEx

    Data Scientist Intern

    • Built Predictive model with 78% accuracy using python to identify the total lbs. of good the client will be sending in the coming six months of his contract
    • Built an ETL pipeline using Abinitio to extract 6.7 million rows of data from multiple enterprise databases and loaded it into Teradata
    • Integrated Tibco Spotfire with Teradata and developed analytical dashboard which depicted performance across sales

  • Quinnox Inc.

    Data Analyst

    • Analyzed data, developed and implemented effective rules-based fraud detection and prevention strategies and reduced ATM/POS/Internet Banking initiated credit/debit card fraudulent transactions by 22%
    • Developed analytical dashboard in tableau to provide actionable customer insights across different customer groups, segments and region

  • Government College of Engineering

    Bachelors of Engineering

    I did my Undergraduation at GCOEJ in India, majoring in Electronics & Telecommunication.


Projects


Take a look at some of my interesting big data / data analytics projects.
Hover to view project description, click to view the details.

  • All
  • Python
  • Deep-Learning
  • Spark
  • NLP
  • Machine Learning
  • Hive
  • Web Scraping
  • Tableau
Pier 1

Pier1 Imports Analytics Competition Winners.


Analyzed 200k+ customer reviews of products using Natural Language Processing to understand the impact of customer sentiments on 37k+ Pier1 home décor products.

Sign Language Interpreter

Sign Language Interpreter - Hackathon Winners


Implemented a sign language interpreter to help deaf people with their daily communication needs using Deep Learning. The project was completed in 24 hours as part of HackUNT-19, the University of North Texas's annual Hackathon.
Anomaly Detection

Anomaly Detection - Data Science Competition


Build different machine learning models and trained them on 6.7 millions records in Google cloud platform to predict the anomoly with 99.99% accuracy.
energy

Energy Efficiency Optimization - Neural Network


Adding cooling load and heating load can define the overall load of the apartment. Studied the trend of overall load and divided it into three classes, low efficient, high efficient and average efficient. Then train a deep learning model to predict the label..
NLP Spam Classifer

NLP Spam Classifier


Developed NLP model in python, which classifies emails into spam/non-spam and deployed it using Flask.
TwitterVRse

Cardiac Arrhythmia Multi-Class Classification


Build Machine Learning models to classify patients into one of the sixteen subclasses, among which one class represents absence of disease and the other fifteen classes represent electrocardiogram records of various subtypes of arrhythmias.
Sign Language Interpreter

LA Restaurant Data Analysis-Informs Analytics Challenge Winners


An analysis of the inspection data of the restaurants in the 88 cities of Los Angeles county. The project was done as part of INFORMS Analytics Challenge at the University of Texas at Dallas.
Web Scraping

Web Scraping watch data


The main aim behind extracting the watch data is to use it and create a watch recommendation engine using Machine Learning. I used beautiful soup to extract data from world of watches website.
PUBG

PUBG Game Data Analysis


An analysis of Player Unknown's Battle Grounds (PUBG) Game Data using Hive and Spark. After seeing the success of PUBG we wanted to understand what are the key factors to win in this game.
US Open

Twitter Sentiment Analysis - US OPEN


Being a sincere tennis follower I wanted to know who is the favourite among all the 2017's US open-Tennis grand slam semi-Finalists. To find that answer I analysed the tweets using sentiment analysis.