Tanaya Kavathekar
202.290.5772 ♦ Washington, DC ♦ tanaya_10@gwu.edu ♦ github.com/Tann10 ♦ Linkedin.com/in/tanayakavathekar
SUMMARY
 Data Scientist with 3 years of experience analyzing, driving insights, designing, and implementing innovative
solutions for Fortune 500 clients
 Adept in big data tools, scaling analytical solutions in production using parallelization, test-driven development,
and object-orientated programming
EDUCATION
The George Washington University Washington, DC
Master of Science, Data Science, (GPA: 4.00) Anticipated May 2021
Relatedcoursework: Machine Learning, Natural LanguageProcessing, Neural Networks, High Performance Computing
University of Pune Pune, India
Bachelorin Engineering, Computer Science graduated with Distinction May 2016
TECHNICAL SKILLS
Programming: Python, R, Spark, Hive, SQL, Django, Shell, Unit Testing, Java, C, C++
Statistical Techniques: TimeSeries Forecasting, Supervised Learning, Unsupervised Learning, Deep Learning
Software: Excel, Git, Microsoft Azure, Microsoft Office Suite, JIRA, Jenkins, AWS, Docker, Keras, Tensorflow, Pytorch
Data Visualization: PowerBI, Tableau, R & Python Visualization Packages
PROFESSIONAL EXPERIENCE
Mu Sigma Business Solutions Pvt. Ltd. Bangalore, India
Decision Scientist Sep 2016 – Jul 2019
Fortune 100, US-based Manufacturer, Supply Chain & Data Science Team
Demand Forecast
 Engineered data pipelines to pull ~2TB data from varied sources in Azure Data Lake to analyze trends and patterns
to build robust volume forecast framework
 Improved accuracy of demand forecast by factor of 8% and achieved 15% improvement in case fill rate for supply
chain division by building ensemble time-series models
Fortune 200, UK-based RetailChain, Data Science & Technology Team
Sales Forecast & Diagnostic Engine
 Increased the forecast accuracy of sales value and volume by 5.6% consumed by Finance Team, benchmarked
against the best legacy system, by implementing ARIMA with a customized seasonal adjustment
 Engineered parallel processing for model building, scoring, and forecasting for ~2500 stores and ~3600 product
groups constituting ~1 TB of data using Hadoop andPySpark technology
 Instituted automated insight tool to monitor forecast coverage and reliability which established the use of
centralized forecast system
Demand Transfer on the Delisting of Products
 Lifted profit margin by 1.3%by predicting demand transfers for delisting products and translating into different
KPIs, by building a union of ensemble models (time-series + randomforest) and business heuristics (22k models)
 Designed User-Interface in Django to drive consumption of the analysis across the organization to enable data-
driven decision making
PROJECTS
Predicting Taxi Tip Amount (Link) Oct 2019
 Implemented multiple regression models, regularization techniques, and decision trees to predict tip amount for
New York Taxi cabs with an accuracy of ~90% using R
Popularity Level Classification (Link) Dec 2019
 Built Grid Search with 7 different algorithms to classify YouTube videos based on the number of views with an
accuracy of 83%, in order to efficiently plan advertisement strategies using Python
LEADERSHIP AND ACCOMPLISHMENTS
International Student Association, GWU, Associate Director Sep 2019 – Present
 Supported international graduatestudents by identifying careerand academic needs
 Liaised with departments to gain support for planned events that enhanced the international student experience
Bhumi NGO, Program Coordinator and Instructor: Designed and executed primary mathematics curriculum in local
government schools Jul 2018 – Jul 2019
Publication: Creating Cloud for Virtual Lab (Research paper), Cyber Times Journal, Nov 2015
Publication: Creating Cloud for Virtual Lab (Implementation paper), IARJSET, May 2016
Received3 Mu Sigma Spot Awards: Creative Problem-Solving Ability; Independently Working on Multiple APIs; Thought
Leadership and Diverse Technical Capability

Tanaya jan 17 Resume

  • 1.
    Tanaya Kavathekar 202.290.5772 ♦Washington, DC ♦ tanaya_10@gwu.edu ♦ github.com/Tann10 ♦ Linkedin.com/in/tanayakavathekar SUMMARY  Data Scientist with 3 years of experience analyzing, driving insights, designing, and implementing innovative solutions for Fortune 500 clients  Adept in big data tools, scaling analytical solutions in production using parallelization, test-driven development, and object-orientated programming EDUCATION The George Washington University Washington, DC Master of Science, Data Science, (GPA: 4.00) Anticipated May 2021 Relatedcoursework: Machine Learning, Natural LanguageProcessing, Neural Networks, High Performance Computing University of Pune Pune, India Bachelorin Engineering, Computer Science graduated with Distinction May 2016 TECHNICAL SKILLS Programming: Python, R, Spark, Hive, SQL, Django, Shell, Unit Testing, Java, C, C++ Statistical Techniques: TimeSeries Forecasting, Supervised Learning, Unsupervised Learning, Deep Learning Software: Excel, Git, Microsoft Azure, Microsoft Office Suite, JIRA, Jenkins, AWS, Docker, Keras, Tensorflow, Pytorch Data Visualization: PowerBI, Tableau, R & Python Visualization Packages PROFESSIONAL EXPERIENCE Mu Sigma Business Solutions Pvt. Ltd. Bangalore, India Decision Scientist Sep 2016 – Jul 2019 Fortune 100, US-based Manufacturer, Supply Chain & Data Science Team Demand Forecast  Engineered data pipelines to pull ~2TB data from varied sources in Azure Data Lake to analyze trends and patterns to build robust volume forecast framework  Improved accuracy of demand forecast by factor of 8% and achieved 15% improvement in case fill rate for supply chain division by building ensemble time-series models Fortune 200, UK-based RetailChain, Data Science & Technology Team Sales Forecast & Diagnostic Engine  Increased the forecast accuracy of sales value and volume by 5.6% consumed by Finance Team, benchmarked against the best legacy system, by implementing ARIMA with a customized seasonal adjustment  Engineered parallel processing for model building, scoring, and forecasting for ~2500 stores and ~3600 product groups constituting ~1 TB of data using Hadoop andPySpark technology  Instituted automated insight tool to monitor forecast coverage and reliability which established the use of centralized forecast system Demand Transfer on the Delisting of Products  Lifted profit margin by 1.3%by predicting demand transfers for delisting products and translating into different KPIs, by building a union of ensemble models (time-series + randomforest) and business heuristics (22k models)  Designed User-Interface in Django to drive consumption of the analysis across the organization to enable data- driven decision making PROJECTS Predicting Taxi Tip Amount (Link) Oct 2019  Implemented multiple regression models, regularization techniques, and decision trees to predict tip amount for New York Taxi cabs with an accuracy of ~90% using R Popularity Level Classification (Link) Dec 2019  Built Grid Search with 7 different algorithms to classify YouTube videos based on the number of views with an accuracy of 83%, in order to efficiently plan advertisement strategies using Python LEADERSHIP AND ACCOMPLISHMENTS International Student Association, GWU, Associate Director Sep 2019 – Present  Supported international graduatestudents by identifying careerand academic needs  Liaised with departments to gain support for planned events that enhanced the international student experience Bhumi NGO, Program Coordinator and Instructor: Designed and executed primary mathematics curriculum in local government schools Jul 2018 – Jul 2019 Publication: Creating Cloud for Virtual Lab (Research paper), Cyber Times Journal, Nov 2015 Publication: Creating Cloud for Virtual Lab (Implementation paper), IARJSET, May 2016 Received3 Mu Sigma Spot Awards: Creative Problem-Solving Ability; Independently Working on Multiple APIs; Thought Leadership and Diverse Technical Capability