Shouvik Sharma Email & Phone Number

California, United States

- Developed and maintained highly scalable ETL pipelines using Databricks and dbt, ensuring data integrity and seamless integration across multiple data sources.- Designed and optimized data workflows, reducing processing times by 30% through the use of Apache Spark and Databricks, resulting in improved data pipeline efficiency.- Orchestrated complex.

- Implemented robust data quality checks using SODA for real-time validation, resulting in a 30% reduction in data discrepancies.- Led end-to-end data transformation workflows with dbt, achieving a 20% improvement in analytics process efficiency.- Extracted insights from marketing data to optimize customer application experience, contributing to a 4%.

Racine, Wisconsin, United States

- Led the development and maintenance of data warehouses on Dremio and Databricks, improving data accessibility and efficiency.- Implemented ELT processes and conducted data analysis using SQL and Python.- Worked on cloud computing services (AWS) to develop scalable and efficient data solutions.- Collaborated with cross-functional teams to implement BI.

Racine, Wisconsin, United States

• Development of predictive analytics and statistical forecasting models utilizing forecasting related systems to improve forecast accuracy and bias reduction
• Formalizing assumptions about how demand forecasts are expected to behave, creating definitions of outliers, developingmethods to systematically identify these outliers, and explaining why they are reasonable or.
• Development of tools to allow process automation, analysis & corrective action implementation by the business.- Forecast Accuracy analysis and corrective action implementation

Schaumburg, Illinois, United States

- Developed and automated data migration pipeline from SQL Server to Snowflake using SnowSQL and SnowPipe, and performed dimensional modeling on the migrated data, further created data dictionary for the technical audience.- Automated ETL processes using Prefect (Python), making it easier to wrangle data sets and reducing time by as much as 40% by.

Schaumburg, Illinois, United States

• Developed statistical models like ARIMA using statsmodels package in Jupyter Notebook,the model achieved an overall accuracy of MAPE 5.96%
• Created interactive dashboard using R Shiny
• Automated the end-to-end model workflow using Azure DevOps and Azure MachineLearning which allows CI/CD architecture

Chicago, Illinois, United States

Chicago, Illinois, United States

• Analyzed sales of multiple departments using a dataset of over 300 features in Python which included KPIs such as transport market indicators, freight movement, and US economic indicators
• Removed multicollinearity from domain knowledge and identified important features with respect to the dependent variable using the Granger Causality test
• Forecasted sales for 2021 using a combination of Statistical (SARIMAX) and Deep Learning Time Series (LSTM and RNN layers)techniques which resulted in a Mean Absolute Percentage Error rate of 8%
• Deployed models by creating an interactive Flask application and created visualization dashboards using Tableau; presented analysis to senior leadership

Mumbai Area, India

1. Developed customer insights for one of India’s largest grocery chains, to assist their marketing team for improving Customer Retentions, Reducing Churn Rate, Campaign Responses, Lift and Incremental revenue using statistical techniques like RFM methodology, Linear Regression and Logistic Regression. 2. Built CLTV & BTYD propensity models using BTYDplus.

Mumbai Area, India

• Led a price forecasting project for Greeksoft Technologies Pvt. Ltd., Mumbai
• Extracted stock price data using NSEpy library which is used to extract historical and real-time data from NSE’s website inpython.
• Created external variables using technical stock indicators for determining the impact on the closing price
• Performed data cleaning and data manipulation for excluding holidays in the stock market
• Built an RNN Neural Network model for Live positional trading using Keras package with Tensorflow in AWS Sagemaker, whereoutputs supplemented Bull Spread Strategy in Options Trading, the developed model architecture.
• Deployed automate end-to-end predictive modeling pipeline using AWS DevOps, where it supported automated daily priceforecasting using the LSTM neural network architecture

Baroda

• Worked as Data Science Intern to automate sample design processes using R software
• Assisted in designing and development of technical architecture for sample design process
• Reduced time required to complete these processes by 25%, thereby helping management to make important decisions faster
• Propose potential research-on-research tests to improve current Nielsen methodologies and improve response andcompliance
• Dive in, and work with our data science team to develop new data-centric products involving new and innovative algorithms
• Classification of store types based on store attributes using Random Forest algorithm in PySpark which resulted in bettersurveying and data collection

Lower Parel

1. Drive acquisition channel of used cars and two-wheelers dealership, by building customer scorecard analysing different parameters affecting repaying capacity2. Identified key metrics by extracting and analysing client’s customer data; gained an understanding of the client’s financial and inventory process for deriving second-order variables, to.

Master'S Degree, Data Science, 3.83

Graduated with Magna Cum Laude Relevant Courses - CS 584 - Machine Learning, CSP 554 - Big Data Technologies, CSP 571 - Data Preparation &.

Masters, Statistics

Relevant Courses 1) Distribution Theory 2) Linear Algebra and Numerical Methods 3) Parametric Inference estimation 4) Probability Theory.

Education record

Education record