Data Python

Nirlep Shah

*********@********.*** 682-***-**** http://www.linkedin.com/in/nirlepshah1 https://github.com/Nirlep-Shah

PROFILE HIGHLIGHTS

Data Science Professional with 2+ years of experience in Healthcare Industry in facilitating executive decision-making for various teams by implementing bigdata technologies and statistical analysis.

EDUCATION

University of Texas at Dallas: Masters Business Analytics Aug 2018 - May 2020

National Institute of Technology: Bachelor of Technology - Industrial Engineering Aug 2012 - May 2016

TECHNICAL SKILLS

AWS Core Services: Compute, Networking, Storage, CDN, Database, VPN, Routing, Application Services

Languages: Python, R, SQL, Java, SAS, XML, MS Office, MongoDB, MS Excel (Match, VLOOKUP), Tableau

Bigdata Technologies: Hive, SQL, Unix, Hadoop, Bash scripting, Java, Python

Machine Learning: K-means, Naïve Bayes, SVM, Logistic Regression, KNN, Random forest, Decision Tree

PROFESSIONAL EXPERIENCE

RailcarRx, Inc. (a PRIMUS sister company) Jul 2019 – Nov 2019

Intern – Jr. Cloud Engineer (AWS)

Study existing RailcarRx technology landscape and understand current application workloads.

Configured cloud watch dashboards, billing alarms and performance alarms to monitor RMS, Insight and CRB applications and do proactive maintenance.

Designed step function to acquire industry data for railcars from multiple web services and score the cars based on the information.

Grouped resources and organized the billing dashboards to understand cost vs usage by customers.

Organized the IAM for the team and setup roles with granular permissions.

OPTUM (United Health Group) Jun 2016 – Jun 2018

Associate Data Scientist Jan 2018 – Jun 2018

Professional Claim Regression: Reduced execution time from 7 days to 90 minutes by converting a predictive model from SAS to PySpark and migrating it to Hadoop environment. Optimized and automated SQL queries and created a process to flag and send fraudulent provider names to vetting team according to business logic.

Physician Behavioral Analysis: Boosted sales by 18% and plunged processing time by 75% by developing multiple interactive dashboards after analyzing PMR data of physicians, using Tableau, SQL

Business Analyst Jun 2016 - Dec 2017

837 Auto Process Development: Developed an automated standalone batch level process in production environment to parse EDI 837 files and store the parsed data into relational databases using Java, Unix, Hive, SQL queries and other bigdata technologies.

UDW-837p Matchup and Missing dates analysis: Devised matchup logic to create UDW-837p database to join two databases and found missing claims and lines not sent by UDW team using Excel and SQL queries.

837I SQL ETL: Designed ETL (Extract-Transform-Load) on SQL for claims data source 837I to convert raw data into relational database, which updated the daily database automatically.

ACADEMIC PROJECTS AND COURSEWORK

Linear Regression (Python): Predicted total number of comments a Facebook post will receive in next H hours using gradient descent as cost and error measures with batch update with 80% accuracy.

Classification Model (Python): Predicted whether the client will subscribe for term deposit or not by implementing 3 machine learning algorithms - Support Vector Machines (SVM), Decision trees and Boosting and compared results to find the best model.

Student Grade Classification (Python): Predicted student’s final grade with 71% accuracy by creating model based on earlier performance and demographic information data

ACHIEVEMENTS AND LEADERSHIP EXPERIENCE

Data Analytics Lead (Enactus UTD): Finding insights from data collected for homeless in US for current project.

SME and Mentor: I was SME for 837 data and mentored young professionals on 837 data.

Recognized for quality for streamlining 837P Process and creation of 837I Parser.

Contact this candidate