Cassie Williams

Data Science Manager @Bombora

Las Vegas, NV, US

Lead the Intent Data Science team, partnering with product and engineering to define and execute the roadmap; align OKRs, create project timelines, and ensure successful delivery and production ready handoff of ML solutionsTranslate customer feedback into actionable research problems, designing data science solutions to improve product accuracy, reduce churn, and enhance customer onboardingDeveloped a patent-pending funnel phase analysis application to identify research trends of closed-won accounts, achieving >85% precision across customers and significantly accelerating the topic selection process; implemented using BigQuery, Vertex AI, time series analysis, and statistical testingDesigned and launched an automated topic selection system using BigQuery, Vertex AI, and LLMs, reducing onboarding time and enabling automation where none previously existedContinuing to develop an end-to-end topic onboarding pipeline with Kubeflow Pipelines and LLMs to streamline onboarding, improve customer experience, increase retention, and accelerate release frequency from 3 times per year to every 6 weeksSupported development of a multi-label classification model using PyTorch and Hugging Face Transformers to enhance real-time data processing and classification for downstream intent signals, achieving an 80% F1 score and improving tagging precision and scalabilityContributed to the development of a disambiguation model and automatic alias generation pipeline using Kubeflow Pipelines, leveraging sentence transformer embeddings and cosine similarity to improve signal quality and entity resolution; achieved 85% precision across evaluated datasets

Penn State University

Master's degree, Applied Statistics

University of Cincinnati

Bachelor's of Science, Mathematics and Actuarial Sciences