I oversee the development of products meant to support data science work at Allstate. I am currently focused on building a product to support model deployment to mobile devices. This involves:- Agile Product Management- Doing Data Science work in support of the product- Communicating Data Scientist needs to developers- Advocating data science to a non-technical audience- Providing technical data science and development support

My work is primarily focused on using Allstate's telematics data to support novel insurance products. I have also been involved in creating the next generation of tools to collect telematics data for Allstate.-Used big data tools (Spark, Sqoop, Impala) to clean and analyze huge datasets-Created a tree-based model to increase the accuracy of hard braking detection-Built pricing model that utilized location and motion data-Built a linear model to detect data quality issues-Wrote code in Python using machine learning and data analysis libraries (numpy,scipy,sklearn,pandas,joblib) to parse and extract information from data-Wrote Scala-Spark code to cluster and analyze terabytes of location data in parallel-Oversaw the development of test mobile apps to further refine models

My work is primarily focused on processing terabytes of location and motion data generated by Allstate vehicles. I am also managing a summer intern.-Patented approaches to detecting braking, turning and collisions on mobile phones-Developed and patented a method to distinguish vehicle movement from phone movement in mobile telematics data

My work is primarily focused on processing hundreds of terabytes location and motion data generated by Allstate vehicles.- Supported efforts to better understand collision detection- Used clustering to better understand customer behavior

This position primarily involved using machine learning techniques to extract meaningful information for physical activity data-Program classifiers, such as Naive Bayesian Classifiers and Decision Trees, to use data from wearable accelerometers to classify user behavior-Write and maintain an Android phone app to collect physical activity level, heart rate, location, and other context data-Analyze power consumption and accuracy of duty cycling methods used to collect data on wearable devices.

-Generate data from Monte Carlo simulations of colloidal nanoparticle self-assembly-Write code in Bash, and Python to analyze large amounts of data-Collaborate across disciplines with engineers and materials scientists