• Understanding the business requirements/needs from Client Big River Services.
• Access download data from different sources text file, excel file, .csv files and databases etc.
• Data upload in S3 Bucket of AWS Cloud Environment and Data Fetching from Redshift Database
• Perform Various Sanity Checks on Data, variable creation, data validation, combining data.
• Carried out simple and complex analysis, useful data extraction and data manipulation as per business requirements.

• Performing exploratory data analysis, identifying study population out of the theoretical population by using population description and finally drawing samples out of it.
• KPI for Retail Sales, Customer Promise Met, Inventory, Operations.
• Using various sampling techniques to draw a sample like Simple Random, Stratified Sampling, etc depending upon the characteristics of the variables.
• Calculating all required statistic on samples i.e., central tendency, variability (standard deviation, z-score)
• Mapping of real-world problem (forecasting of sales, optimising operation efficiency, minimising cost, customer churning etc. into ML problem)
• Building machine learning models Linear Regression, Logistic Regression, CART, Random Forest, SVM, KNN, Kmeans, Classifier(xgboost) etc. depending upon the business requirements
• NLP (Text Mining Project for Customers Sentiments)

• Retail, Operation KPI Calculations & Presentation, Sales Analysis, Customer Segmentation (RFM), Forecasting, Inventory Optimisation, Operation Optimisation (ETL, SQL, EXCEL, RedShift, AWS), Data Managemen

• Understanding the business requirements/needs from Client.
• Access to data from different sources text file, excel file, .csv files and databases etc.
• Perform Various Sanity Checks on Data, variable creation, data validation, combining data.
• Carried out simple and complex analysis, useful data extraction/manipulation as per business requirements.

• Performing exploratory data analysis, identifying study population out of the theoretical population by using population description and finally drawing samples out of it.
• Using various sampling techniques to draw a sample like Simple Random, Stratified Sampling, etc depending upon the characteristics of the variables.
• Calculating all required statistic on samples i.e., central tendency, variability (standard deviation, z-score).
• Building machine learning models Linear Regression, Logistic Regression, CART, Random Forest, SVM, KNN, K-means etc. depending upon the business requirements.
• Mapping of real-world problem (forecasting of sales, optimising operation efficiency minimising cost, customer churning etc. into ML problem.
• Developing ML codes on python using numpy, pandas, matplotlib, scikit
• Credit Risk Modelling using ML, Sales Forecasting, Customer Segmentation (RFM), Inventory Optimisation, Operation Optimisation using ML algorithms.

• Using various sampling techniques to draw a sample like Simple Random, Stratified Sampling, etc depending upon the characteristics of the variables.
• Calculating all required statistic on samples (central tendency, variability, standard deviation, z-score)
• Building various Machine Learning models using Python like Linear Regression, Logistic Regression Random Forest, SVM, KNN etc. depending upon the business requirements.
• Analysing data, drawing conclusions & developing recommendations.
• Developing ML codes on python using numpy, pandas, matplotlib, scikit-learn and park them on GIT for version control.
• Customer Segmentation (RFM), Credit Risk Modelling, Sales Forecasting, Customer Churn Forecasting

Daily Quarterly/Yearly/Reporting:

• Daily CDR Statistics Summary Reports of 24 Hours, Percentage of New Customers and Churned Customers, Calls to International Destinations in decreasing order, Average Revenue per User (ARPU) for last week, Average Revenue per User Segments per Day, Total Customers and their percentage and Bill Amount who has not paid last bills.
• % of Customer who are paying bills on time (On or Before Due Date) and After Due Date.
• Total No of Customers whose Bill Amount is 50 percent of the Total Bill Amount.
• Monthly CDR Statistics Summary Reports, Total Customers and Bill Amount who has not paid last bills.
• Customer counts based on region, gender and age using particular products of the company.
• Highest Revenue and Lowest Revenue from each Customer Segments, Yearly CDR Statistics Summary Reports, Percentage increase or decrease of number of Usages, Percentage increase or decrease of number of Customers, % Increase or decrease of Average Revenue per Customer with respect previous years, % Increase or decrease of Average Revenue per Customer with respect previous Quarter.

Multiple Projects:

NetMind - Fraud Management System (Implementations across multiple clients in US, Europe, Australia, Canada and other countries.