Shashank Jain Email & Phone Number

Washington, District Of Columbia, Us

Bloomington, In, Us

Washington, District Of Columbia, Us

New York, Ny, Us

Applied machine learning algorithms (Regression, K-Means Clustering, Logistics Regression, distance-based matching) on panel data such as, TV viewing data, media data, PC, mobile and tablet usage data, etc. for major Broadcasting and FMCG. Migrated data science models from SAS to distributed framework over in clouds to make them scalable and fast, which helped to boost speed by 70% and reduced the manual efforts by 95%. Implemented continuous integration and continuous delivery with unit tests for various projects. All included data quality checks which helped in identifying data issues before the processes begins. Extracted datasets for fusion processes from Spark using Python/SAS to perform data manipulation for imputation, recoding, and trimming of data to prepare data for fusion engine. Performed preliminary data analysis and visualization to identify important variables between the datasets to be fused by analyzing penetration, correlation and distribution using python packages such as StatsModels, sklearn, keras and Plotly and implemented dimensionality reduction techniques to reduce linking variables if necessary. Incorporated opensource tool Apache Superset for visualization in the team and migrated 13 different reports. Designed DAGs for Machine learning models and Data pipelines using Airflow. Worked on optimization and calibration problem related to surveys & site census using cvxpy python package. Optimized spark jobs resulting in 30% reduction in run time.

Teaneck, New Jersey, Us

Developed queries in Hive, Spark and Oracle to perform ETL operations on the wide variety of data such as click streaming, sales, customer and product catalog. Automated ETL jobs error handling by writing code to take scenario-based action, which made process robust and saved 10% of total run time. Designed Tableau dashboard and reports for different email campaigns metrics such as email clicks, conversion, and bounce rate which helped data science team to evaluate their models without manually pulling data. Maintained the data pipelines from Kafka stream to team’s data warehouse. Kept the data pipeline up and running in production when ETL processes were executing by fixing bugs and issues in tight SLAs.

Master Of Science - Ms, Data Science

Nanodegree, Machine Learning Engineer

Bachelor Of Engineering (B.E.), Electronics And Communications Engineering

Education record