Meet the Data Science Team

The Data Science team at The General® was created mid-2016 and has since grown to include two Data Scientists, and two Machine Learning Engineers, and a Data Science Manager. We wanted to take a minute to introduce you to the team, and briefly share some of the projects or areas of the organization we are each focusing on.

We created “radar charts” for each team member to help visually represent our areas of expertise. While some of the categories are self-explanatory, we highlighted specific subtopics we are including in the more nuanced terms:

RADAR CHART CATEGORIES

DATA SCIENTISTS

  • Data Engineering - data wrangling & manipulation, SQL, ETLs

  • Engineering - deploying data science models into production

  • Deep Learning/AI - NLP, neural networks

  • Machine Learning - supervised and unsupervised learning techniques, decision trees, clustering, OCR algorithms

  • Statistics - GLMs, survival analysis, forecasting, DoX

  • Maths/OR

  • Data Visualization

  • Business Intuition - domain knowledge, strategic thinking, financial literacy, management and leadership

  • Communicator/Storyteller

MACHINE LEARNING ENGINEERS

  • Software Design - code structure, public interfaces, design patterns, maintainability

  • Modeling Knowledge - knowledge to transition data science modeling (GLMs, ML, DL, NLP) into production, including functional and non-functional requirements

  • Back-end Web Development - REST, microservices, caches

  • Model Engineering Pipelines - model training, complex batch prediction pipelines, model performance validation

  • Data Science Frameworks - Scikit-learn, pandas, PyTorch, Keras, h2o

  • Documentation - maintaining a readable code base others can understand, configuring and running an app, doc strings

  • Software Testing - Unit Testing, Integration Testing, Functional Testing, Performance Testing, TDD

  • Software Development Life Cycle - overall comfort with SDLC

  • Data Engineering (ETL) - moving data from A to Z with transformations

DATA SCIENCE TEAM

John Burke, Data Scientist
John joined The General in August 2018 and has been learning the business and familiarizing himself with our many databases. His first project used glmnet in R to create a new symbol set to more appropriately access risk cohorts for specific vehicle makes and models. John’s next efforts will align with our Marketing team. As you may have heard, we recently welcomed Bobby Bones into The General family and we are excited to learn more about how he is connecting with our customers.

john.png

Tim Dobbins, Data Scientist
Tim has been with The General since April 2017. What originally started as an R&D project investigating how our customers were chatting with our customer service representatives, quickly turned into a deep understanding of text analytics. Tim used that to springboard into NLP and has since become our resident expert, applying that area of expertise in various claims projects including a model predicting attorney involvement.

tim.png

Mike Kehayes, Machine Learning Engineer
Mike joined The General in 2017 as part of the Data Warehouse team before transitioning to the Data Science team in the spring of 2018 as the team’s first engineer. He led the effort in creating the company’s first real-time bidding API, allowing us to collaborate with a vendor to support bidding on potential leads in real-time. His attention is currently on other model deployments and working with engineers both on the team and other departments as we continue to evolve and figure out new approaches for model deployments.

Mike.png

Jack Pitts, Machine Learning Engineer
Jack is officially the newest member of the Data Science team. Originally having joined the company on another team (starting to see a pattern yet?), he approached us and started dabbling on the engineering side of the house. Having a natural ability to pick it up quickly, he spent much of his time in 2018 assisting the team with various projects and we were able to make it official in February 2019. Jack is currently concentrating on automated rebasing of model factors and investigating how to implement model monitoring (feature drift, anomaly detection) for production models.

jack.png

Chris Morgan, Data Science Manager
Chris joined The General at the beginning of 2017 as a Data Scientist. In his time with the company his main efforts have been in the marketing space and customer behavioral analytics, be it forecasting, survival analysis, or predicting how to efficiently communicate with customers. His first love is statistics and he will always enjoy a good regression model, but lately his focus has shifted to leading the team and defining data science strategy within the organization.

chris.png