Data Scientist Senior

Priyadarshini selvam

SENIOR DATA SCIENTIST

+1-445-***-**** *******************.********@*****.*** Herndon, VA, USA

https://www.linkedin.com/in/priyadarshini-selvam-a14168119

Career Summary

As a data scientist with close to 8 years of experience, I work closely with the business to handle real world problems that are data driven and propose valuable solutions, predictions and statistical recommendations using my technical competencies and value-based skills, thereby prescribing crucial insights for the business.

Expert in handling data in cloud platform like Snowflakes, AWS, Google Cloud platform

Strong SQL query knowledge and programming expert in python, R and SAS.

Expert in building workflows and Data pipelines in Airflow, Snaplogic for managing the ETL Process and Data Modeling on AWS, PySpark, Hadoop, Snowflake & Kafka

Individual contributor on tools like Databricks, AWS Sagemaker, AWS Data pipeline,Informatia,AWS Lambda, Snaplogic, Snowpipe, PySpark, Airflow, and reporting solution using PowerBI, Tableau

Mentored and developed the capabilities and organizational knowledge of junior associates.

POC and Research implementation of Gen AI into the industry use cases also using Microsoft copilot, Power Virtual Agents etc

Business domain knowledge in Manufacturing, Marketing and finance industrial verticals

Education and Certification

Masters degree in Computer Science College of Engineering Guindy, Anna University

May 2017

Analytics Professional Development Program (18 months program)

Dec 2018

Blockchain Use case & Architecture from NPTEL

Apr 2019

Introduction to Data Science (IIT Madras) – Caterpillar Continued Learning Program

Dec 2019

Marketing Analytics – from NPTEL, certified by IIT Roorkee

Dec 2020

Scrum Master Certified (SMC) – Scrum Alliance

Oct 2021

Technical Skills

Tools & Languages

AWS (Sagemaker, Lambda,Glue,EC2), Python, R, SAS(Both Advanced & BASE SAS), SQL, C, DataBricks, Apache Spark, Streamlit, Apache Kafka, Hadoop, Informatica, MongoDB, Microsoft Copilot ( PVA), pyTorch, Airflow

Visualization

Power BI, Tableau, HTML, CSS, JavaScript, Google Looker Studio

Domain

NLP (Natural Language processing), Image Processing

GenAI

LLM, LangChain

Cloud Platform

Snowflakes, Google Cloud Platform(GCP), Google Big Query, AWS Cloud Practitioner for Data Science

Work Experience

Data Analyst Caterpillar Inc. Feb 2025 – Present

As a data analyst for the conditioning monitoring platform, worked closely in analysing the customer fleet to help troubleshoot issues in the condition monitoring platform.

Analyzed the data from the customer Telematics device to learn and identify the root cause of the reporting status through Analysis tools in Snowflake, SQL and Presented the same in Power BI

Reported back to the Business on the Connectivity rate and the value story of the group to help track the opportunity

Data Scientist Kaytics Inc. Nov 2024 – Feb 2025

Prototyped a SAAS product which delivers Marketing analytics solutions to a potential marketing agency which help budget decisions, digital and media marketing allocations. The product build in python and Azure helps in optimising the marketing budget spend and also simulates the projected ROI % for the next quarters

Built a Digital media marketing attribution model for a CPG Client using python

Revamped enterprise visualization for faster and quicker insights in Power BI Report server

Customer segmentation - Aims at identifying the customer group based on the customer survey comments using Topic Modelling in Text Analytics and NLP Market Mix modelling :

Contributed in building a marketing mix modelling in identifying which category or channel is most effective in conversion and based on which allocation are performed.

Data Scientist Caterpillar Inc. Chicago, IL Jul 2017- Sep 2024

Manufacturing Analytics (Enterprise Analytics group):

Worked on the Equipment Care Advisor ( ECA )and telematics group and contributed in prediction of the machine failure ( RUL-Remaining Useful Life)

Product Link - Alert analytics ( Which works on applications to users on timely alerts etc) - Timely reminders for Scheduled Oil sampling

Visionlink - impact of marketing on the various subscriptions of visionlink ( basic to advanced )

Marketing Analytics -

Implemented the Marketing to Sales attribution model, to help attribute a sale to an online marketing campaign /email/ website click, thereby justifying the ROI on the Marketing spend. Enhanced the model through a mix of data & Multi touch attribution algorithms thereby improving the attribution from 10% to 12%.

Developed a Market Mix Model for optimizing the spend in various channels with a simulation tool to visualize the impact on sale. Participated in a Workshop in North Carolina and Illinois with the Business team to help improve the model. Built a validation matrix to help access the accuracy of the developed model and improved the accuracy by 5%.

Developed the Ready to Buy – List of customers who are at the potential to buy the next product based on an extensive customer scoring and targeting which resulted in highest conversion rates and a buy in from many business teams within the organization ( improved the recall by 2% and accessed and added two high impact variables).

Built a chat bot to integrate with the Power BI dashboards to help interactive user search and Q&A with the data in the dashboard using Microsoft Power Virtual Agent and Microsoft Power Automate

Developed a Weighted fuzzy match algorithm for customer data match based on which a White paper titled ‘Data Match algorithm using Weighted Averages’ was written which improved the match % from 30% to 45%

Text Analytics: -

Have Analyzed the Customer Survey Comments, to identify the potential areas of improvements provided as suggestion by the customer and highlight the Key Topics of appreciation and Sentiment Analysis on the customer’s perception

Implemented a Part distribution Dashboard in power BI, which tries to identify any part related information from the dealer service network complaint text and tries to correlate any new complaints to the existing list of potential issues thereby providing visibility to the business.

Social Media Analytics: -

Have experience handling the unstructured data from social media platforms to Topic Model the area of concern a customer is more frequently talking about.

ROI (Return of Investment Analysis) and Campaign effectiveness analysis for special media campaigns run as a part of Online Marketing.

Finance Analytics -

Performed Financial Data Analytics for the Enterprise Financial Shared Services (EFSS) team of Caterpillar to identify and aid in the invoices processed every month, detect delays, and help improve the efficiency of the process by forecasting possible anomalies. Also visited the team in Belfast, Northern Ireland in person for consulting (Reduced the processing by 4X times).

Worked with General Ledger accounting, Capital assets, and Payroll teams in areas like data quality assessment, reconciliation, outlier analysis, trends/patterns on invoicing process thereby improving the process quality, velocity and significantly reduced the manual hours.

Project Management:

Certified Scrum Master handled several Analytics projects in the Agile Scrum Methodology.

Lead advanced analytics projects end to end, right from scope definition, stakeholder management, key milestone definition, internal brainstorming of ideas, feasibility analysis and delivery management.

Managed VSTS (Visual Studio Project track) completely in defining and assigning the various activities for the team.

Automation and Reusability:

Automated several repeated tasks to reduce time and manual intervention by analyzing and optimizing several long running processes using SAS and python.

Developed functions/ macros for code reusability purposes of repeated tasks / activities.

Other responsibilities undertaken -

Part of the recruitment process for hiring of the best talents into the Marketing Analytics group.

Mentored associates within the team on SAS and python skills.

Organized “Marketing & Brand- Getting to know” networking session with the leaders for business understanding within the marketing and brand team.

Completed the 18 Months Analytics Profession Development Program as a part of the Information Analytics Group, Caterpillar India.

College Intern Caterpillar Inc. Dec 2016- Jun 2017

Developed a proof of concept and implemented an end to end product on Complex Event Processing (CEP) using Apache spark streaming on the near real time unstructured data as a part of thesis for the University.

Engineering Intern Red Black tree technologies Jun 2015- Nov 2015

Developed Android applications for Event Management firm.

Worked on the architecture and development of a complete web application using node.js

Contact this candidate