Personal details

Arul B.

Timezone: Pacific Time (US & Canada) (UTC-7)

Summary

Data Scientist / Applied Machine Learning Engineer with more than five years of experience in exploring, analyzing, and researching financial, real-estate and user behavior data to procure insights, prescribe recommendations, build models, design experiments and deploy scalable machine learning applications.

Technical skills

Data Science・4 yrs Deep Learning・3 yrs Python・4 yrs R・4 yrs Apache Spark・3 yrs

Work Experience

Data Scientist & Applied Machine Learning Engineer

Realtor.com | Jan 2018 - Present

Python

Machine Learning

Statistics

Data Science

Deep Learning

PyTorch

• Created scalable and optimized SQL queries to retrieve and aggregate terabytes of data from AWS Athena data lakes • Performed Consumer Behavior Modelling using Tree-based Model Interpretation and singled out key variables impacting Consumer Churn • Trained and deployed PyTorch CNN models(ResNext, ResNet, VGG16) using AWS SageMaker

Senior Developer (Business Intelligence & Data Analytics)

Bank of New York Mellon | May 2013 - Jul 2017

Python

Java

Angular

Spring

• Drove data analytic applications that involve metric reporting of financial and performance metrics of the firm. • Mined and analyzed massive sets of Workforce Analytics data to identify the key trends of employee engagement for increasing operational efficiency and workforce performance. • Analysed the trends of telecom, travel, infrastructure, and technology expenses incurred by the firm and predicted actionable insights to reduce the overall expense quotient of the firm. • Developed queries and maintained database architecture for My Dashboard, an application that is used by the employees and clients across the worldwide branches of BNY Mellon. • Won the BNY - Best of Class Award for successfully developing My Dashboard application and delivering it to numerous clients across the world. • Knowledge-engineered actionable insights by mining financial and performance data to ensure efficient operation and functioning of the proper business sectors. • Took care of End to End Delivery, Technical Support, and Documentation.

Personal Projects

Rival Check Cross Correlator for Locating Strategic Defense Bases Using Supervised Learning

2017

Machine Learning

AI (artificial intelligence)

The need of machine learning in the defence planning and strategies is increasing day by day due to the increasing amount of breaches and decimations caused by terrorist forces. A myriad of military bases, temporary campaigns, base camps etc. are being targeted and attacked by several terrorist forces. The common problem in the warfare and tumultuous international borders is the frequent and violent intrusion and breaches upon the temporary / permanent military and army bases. Though they are successful in their individual task to identify the safest or the effective base, a combined location that embraces both effectiveness and vulnerability is invalid using a present analyzing and classification technology. This problem is due to the presence of collinearity between the parameters that determine both effectiveness and vulnerability. A military base location can be both effective and vulnerable at the same time, a location that does not provide sufficient effectiveness to perform military operation. To combat this problem, in this paper we propose an algorithm that identifies the two rival parameters (effectiveness and vulnerability) and cross correlates them one by one for checking collinearity between them. Additionally, after identifying the collinear combinations, the Rival Check Cross Correlation Algorithm eliminates those collinear combinations, thereby providing unambiguous combinations of effective variables.