Aman Mathur Email & Phone Number

Menlo Park, California, United States

Redwood City, California, United States

Culver City, California

Member Churn Prediction• Built an in-house member churn predictor tool using Stochastic Gradient Boosting algorithm with implementation in R. The tool was deployed using R Shiny on shinyapp server.• This tool helped our Member Success Team in prioritizing conversations with members with highest churn potentialLead Scoring Model• Worked in collaboration with a Senior Data Scientist to build a Lead Scoring model based on historical data using Random Forest. The model provides… Show more Member Churn Prediction• Built an in-house member churn predictor tool using Stochastic Gradient Boosting algorithm with implementation in R. The tool was deployed using R Shiny on shinyapp server.• This tool helped our Member Success Team in prioritizing conversations with members with highest churn potentialLead Scoring Model• Worked in collaboration with a Senior Data Scientist to build a Lead Scoring model based on historical data using Random Forest. The model provides a probability score of conversion to a Sale based on the data provided in the Lead forms• Deployed this model on AWS Lambda service and used Marketo Webhooks make API calls to the Lambda service. The model scores the lead within seconds from the time it enters in our system. Employed as the First Data Scientist at the company. Helped build a data driven culture and leverage new technologies. Show less

Bengaluru Area, India

Weekly Call Volume Forecasting Model with Dual Seasonality – R, R Shiny- Designed a R based forecasting model with dual seasonality data for call center which decreased the prediction error by 15% Grouping Customers for Product Recommendation - Python, Apache Spark, Scikit-Learn, SQL- Implemented a python program in pyspark for Locality Sensitive Hashing algorithm to group similar customers based on online purchase behavior for product recommendation. This lead to increase in… Show more Weekly Call Volume Forecasting Model with Dual Seasonality – R, R Shiny- Designed a R based forecasting model with dual seasonality data for call center which decreased the prediction error by 15% Grouping Customers for Product Recommendation - Python, Apache Spark, Scikit-Learn, SQL- Implemented a python program in pyspark for Locality Sensitive Hashing algorithm to group similar customers based on online purchase behavior for product recommendation. This lead to increase in click through rates by almost 20% Show less

Masters Degree, Computer Science With Specialization In Data Science

Artificial Intelligence, A

Bachelor Of Technology (Btech), Computer Science And Engineering

Activities and Societies: Drama Team, Co-ordinator of Technical Festival.