Aug 2021 – May 2025 · 3 yrs 9 mos · Bengaluru, Karnataka, India · On-site

• Lead the development and implementation of a new product which includes a predictive model (a Xgboost Model with Accelerated Failure Time Loss function) and a TBATS Forecasting framework to predict the optimal date of maintenance on Conveyors for Amazon US Fulfilment Centers. The process generates automated maintenance work orders , reducing 89K labor hours across 259 sites saving cost of $4.4MM.

Jan 2018 – Aug 2021 · 3 yrs 7 mos · Bangalore

• Worked as Assistant Manager responsible for Advanced Analytics development that includes building propensity models ,statistical segmentations using different statistical as well as advanced machine learning techniques thereby supporting the business of Private Label Credit Cards of a major US home improvement client on the behalf of the Company's Retail Service team.
• Analytical Methods used :Random Forests,Extreme Gradient Boosting Machines, CHAID and CART Decision Trees, Logistic Regression,Gaussian Mixture Models, KMeans Clustering Algorithm.
• Also developed a robust framework for developing Machine Learning (Supervised and Unsupervised model) which includes automated data pull codes for pulling variables from different data sources used for creation of several model variables and automated codes for Xgboost models (in R) and K Means Clustering algorithm (in SAS).
• Worked on transformation of entire process from SAS to EAP Big Data platforms with the leadership team and building capabilities for PySpark, Hadoop, Revo R and Hive environments.
• Built codes on Hive to migrate processes on behalf of the entire business vertical and automating modelling codes using H2o.Ai interface (in R) after connecting Revo R with Hive tables.
• Softwares used: SAS, R programming, Python, MS Office , Revo R, PySpark, Hive, Hadoop.

Jun 2016 – Jan 2018 · 1 yr 7 mos · Bangalore

• Worked as data scientist and supporting the advanced analytics team by pooling useful insights from data by exploration and building advanced predictive models using different statistical , machine learning and data mining techniques specially for Retail ,FMCG ,Human Resource and different cross domain clients thereby supporting the business teams.
• Techniques used: Linear Regression ,Logistic Regression ,Decision Trees, Random Forest ,Extreme Gradient Boosting, Market Basket Analysis techniques like Apriori algorithm ,Sequential Market Basket Analysis using Spade Algorithm ,Holt-Winters Method, Pricing using Price Elasticity, K-Means Clustering,Minibatch K-Means Clustering,Recommendation Engine using Collaborative Filtering etc
• Softwares used: R Programming,SAS, SQL, R-Shiny,MS-Office

Jul 2015 – Jun 2016 · 11 mos · Bengaluru Area, India

• Data Understanding:Performing Exploratory Data Analysis to understand the effects of different types of marketing variables on Sales of different FMCG and luxuary goods under most eminent brands of India and foreign markets.
• Modelling:Statistical Modelling using Statistical techniques like Regression,logistic Regression,Time Series Modelling to get deeper insights on the effects of different marketing variables on sales.
• Post Modeling:Providing market strategies by Conducting Deepdive, Optimisation,Calculating Elasticities, Return On Invesment, creating Response Curves after building insightful models.
• Techniques used: Regression Analysis, Time Series Forecasting Techniques like ARIMAX ,Exponential Smoothing etc, Logistic Regression ,Kalman Filter, Vector Auto Regression ,Bayessian Markov Chain Monte Carlo Technique(MCMC technique).
• Softwares used : R Programming,SAS Programming, GRETL, WinBUGS,Genie and Smilie, Bayesia Lab ,MS-Excel