o Served as Lead Data Scientist and Project Manager for a proof-of-concept demonstration of an NLP-based machine learning product (Expertise Inference) to determine distributions and levels of technical and professional skills within McKinsey & Company. Trained the machine learning model and discussed the results with client leaders.
o Refactored the Expertise Inference product from a proof-of-concept implementation into a fully object-oriented architecture for the purposes of scalable, containerized, API-initiated deployments within the IT infrastructure of a major pharmaceutical company.
o Optimized and mapped out conversation flows for chatbots to facilitate the rapid deployment of additional instances for a major health insurance company’s automated customer service chatbots.
o Utilized PySpark to analyze a database of over 3 billion records from an industry-leading consumer mortgage company to find opportunities for targeted mortgage sales, thereby increasing revenue.
o Worked start-to-finish on an Expertise Inference implementation to determine visible and hidden skills for the IT department of a major global airline. Performed data scoping, data collection, preparation, cleaning, all steps of model training as well as project management.
o Analyzed and recovered a large amount of sensitive client data that had been left unusable by prior error, then migrated it with proper protocols for personally identifiable information.
o Created data streaming pipelines and applications to showcase advantages of Python, open-source technologies such as Spark and Kudu and the containerizing of applications with Docker, demonstrated these pipelines to both a multi-company team as well as a large government client.
