Nov 2021 – Mar 2023 · 1 yr 4 mos · Pune, Maharashtra, India

Aug 2020 – Nov 2021 · 1 yr 3 mos · Pune, Maharashtra, India

May 2019 – Jul 2019 · 2 mos · Mumbai Area, India

• Conducted extensive research on UPI frauds and it's modus operandi to identify the inherent gaps in widely used PSP's from the fraud perspective
• Enhanced the framework of fraud mitigation by incorporating the Random Forest algorithm in Big Data environment to ensure negligible lag in alerts
• Recommended 11 new rules based on transaction, registration and demographic variables for effective monitoring and reduction in false positives
• Used SAS Enterprise Miner and R-Studio for modeling the UPI fraud and further deploying it into big data lake using Apache Spark environment

Apr 2018 – Jul 2018 · 3 mos · Nashik

• Demand Forecasting and Inventory Management for Cost Optimization
• Remodeled the existing sales forecasting model at each SKU and regional level for the accurate computation of production as well as transportation cost
• Programmed the model by ensembling the Time Series algorithms viz. ARIMA and ETS and regression algorithms viz. XGBoost and Random Forest
• Reduced the sales forecasting error from more than 50% to within 20% by incorporating various agronomical variables and governmental constraints
• Built a predictive model in R-Studio to predict the optimum level of inventories at SKU level by deploying machine learning and classical methods
• Implemented the model in two manufacturing hubs to achieve the optimal usage of warehousing space in-place with an additional 14% cost saving

Dec 2017 – Feb 2018 · 2 mos · Ranchi Area, India

• Risk Management Strategies for Agricultural Supply Chain
• Conceptualized the strategic model using various supply chain constraints of perishable agricultural products for maximizing the planner's expected profit
• Formulated the model into a Mixed Integer Quadratic Program (MIQP) in IBM Ilog CPLEX software using non-convex MIQP solver to yield the solutions
• Introduced the concept of stochastic demand and price by assuming them to be following Uniform and Normal probability distributions respectively
• Proposed the extension of the model by suggesting the concept of discount pricing for the products after attaining a certain level of perishability

Jul 2017 – Apr 2019 · 1 yr 9 mos · IIT Kharagpur

May 2017 – Jul 2017 · 2 mos · Kolkata Area, India

• Supply Chain Optimization of a leading lead-acid battery manufacturer in India
• Assisted Supply-Chain Management team in implementing pull-based replenishment of the demand for a leading lead-acid battery manufacturer in India
• Designed the monthly demand forecasting framework in R-Studio based on Intermittent and Non-Intermittent Time-Series modeling algorithms
• Optimized the system of freight movements to various warehouses using IBM-Ilog Cplex software, which resulted in the considerable cost savings
• Analyzed the Stock Out Index and Days on Hand for 4 regions to further improve and modify sales-forecasting using Plan Visage and Microsoft Excel

Dec 2016 – Jan 2017 · 1 mo · Kharagpur Area, India

• Annual Yield Forecasting of Crops
• Used Beautiful Soup and Scrapy packages to scrape weather parameters and imputed the missing values using K Nearest Neighbours(KNN) algorithm
• Implemented the concept of Binary Humidity Factor and Growing Degree Days which became the most significant variables by attaining highest weights
• Developed the yield forecasting model in R-Studio using Feed-Forward Neural-Network algorithm and achieved an accuracy of over 90.0%