P.Praveen Kumar

(Data Scientist)

Professional Summary

•Software Developer with 4.8 Years of Experience in Data Science and Java.

•Genpact certified Data Scientist with 3.8 Years of Experience in Machine Learning.

•Achievements include creating data regression models to predict company stock prices with 25%

more accuracy than historical average.

•Highly skilled in Machine learning, data visualization and Data Analytics.

•Good knowledge in Core Java, J2EE, Spring, MuleSoft (API).

•Worked on both Supervised and Unsupervised learning models such as Linear, Logistic regression Random Forest, KNN using Python.

•Statistical modelling techniques like Hypothesis testing, Z testing, ANOVA.

•Performed Exploratory data analysis for complex data and files.

•Good knowledge in Deep learning and Natural Language Processing.

•Have done Pro-degree in Data science.

Work experience

TCS – Data Scientist 09/2020 – 09/2021

Project: Hartford Insurance

Worked in creating a customized python package (automata) which helps to extract, transform and validate company’s internal data and also has features like storing the model metadata into oracle and snowflake tables using PySpark.

Worked with data filtering, EDA process and CI/CD pipeline.

Visualizing the insights of the customer’s policy/insurance data’s for every quarter.

CI Engine which is used to analyses and compare the quotes data of the companies in order to give

better policy rate for the customers.

Automated the scripts using autosys which collects the latest data from hive and generate reports.

Actively engaged in quantitative analysis of sophisticated modelling to address business needs.

Worked on Virtual Environment for creating packages and customized kernels.

Selecting the better model using MLFlow and Model Registry

TCS – Data Scientist 07/2018 – 07/2019

Project: SPRINT DIGITAL TELECOM (SERVICE PLATFORM)

Analysing the Storage space usage for snowflake services for every quarter.

Approached Time Series analysis which helps to identity the trends in our compute habits.

Trend Analysis for days/times when resources are used infrequently and limit the compute nodes

during slow times. Drill down into databases, schemas and query types to track adoption across the organization with user adoption.

Analysing the monthly average for each individual project and usage over time which lies under

cost per limitation of business.

Tuning the predictive churn user activity from past and detecting which customers are likely to cancel the subscription to a service based on their usage. Two class Logistic and Boosted Decision Tree.

Geetham Software Pvt Ltd – Junior Data Scientist 07/2017 – 06/2018

Project: TIME Dataset / Aadhaar Dataset

Collected, studied and interpreted large datasets from different sources for the company and cleansed

it for further analysis.

Analysed the sales trends of the company and helped to predict their requirements for the upcoming sales. Reported healthy and unhealthy months of the company in ordered to improve their revenue.

Suggested to determine what actions to take for targeting the right area of customers in Aadhaar.

Packages used are Numpy, pandas, Matplotlib, sklearn, Keras.

Developed and implemented new forecasting models which increased company productivity and efficiency

Implemented Sentimental Analysis for Aadhaar using NLP (NLTK).

Visualized the data using Tableau. Also got customer appreciations for the reports.

Artificial Neural Networks

Developed a POC model in ANN

Based on the track of six months of the randomly picked customers, have collected those customers

who left and stayed in the bank.

Predictive model which will say which of their customers are at highest risk of leaving the bank

Geetham Software Pvt Ltd – MuleSoft Developer 01/2016 – 06/2017

Project : Transport Administration(TIME-Transport Income Made Easy)

Worked on development of RAML 0.8 using Anypoint exchange.

Integrating the layer for the Module application using Mule Soft.

Developed flow for a set of messages to be converted into the canonical format

MULE EXPRESSION LANGUAGE (MEL) is used for development.

Continuous integration and continuous development using Jenkins.

Implemented development of Error framework for the service layer.

Worked in various connectors, transformers and Listeners.

Bugging out the developed modules using Munit test cases for the integrations.

Profoundly worked on DATA WEAVE EXPRESSION Language.

Skills

Programming Language

Java, Spring, MuleSoft

Data Analytics

Python, R, Data Science, Machine Learning, Deep Learning, PySpark

Databases

SQL, MSQL, Oracle, SnowFlake, Hive

IDE

Eclipse, Anypoint Studio, Anaconda, Jupyter Notebook, Spyder, Colab

Frameworks

Collections

Repository

Git, Bitbucket

Operating Systems

Windows, Unix, Putty

Visualization Tools

Tableau, Excel

Design Patterns

Singleton

Project Tracking Tools

JIRA, Rally

Web Services

RESTful, Postman

Education

St. Joseph’s College

Of Engineering Chennai

Electronics and Communication Engineering Graduated with Aggregate of 56%

09/2011 -06/2015

Vidhya Vikas Higher

12th – State Board

09/2009 -06/2011

Sec school

Graduated with Aggregate of 70%

Petit Seminaire

10th – State Board

09/1997 -06/2009

Pondicherry

Graduated with Aggregate of 77%

Profile

Date of birth:

01/06/1994

Nationality:

Indian

Address:

No: 130, Mahaveer Nagar, lawspet, Pondicherry 600508, India

Phone number:

822*******

Email address:

ado5kj@r.postjobfree.com

LinkedIn:

linkedin.com/in/praveen-kumar-a1b949bb

GitHub:

https://github.com/Praveen8/datascience

Date: P.Praveen Kumar

Contact this candidate