Senior Data Scientist

Total Wine & More

About The Position

Total Wine & More is seeking a Senior Data Scientist to join our Technology team in Bethesda, Maryland. As a Senior Data Scientist, you will play a pivotal role in designing, developing, and deploying machine learning and AI solutions that drive strategic decision-making and operational efficiency across Total Wine & More business. You will be responsible for supporting the full lifecycle of machine learning and AI development—from initial ideation and business problem framing through model development, deployment, and ongoing performance monitoring. This role requires a strong foundation in data science, with a deep interest in learning about production-grade ML systems, and a proactive approach to translating business needs into technical solutions. You will be expected to act independently to deliver high-impact solutions, taking ownership of projects from concept to execution with minimal oversight. You will report to the Sr. Director, Data Science.

Requirements

Bachelor's Degree in Computer Science, Mathematics, Statistics, Economics or related field required or equivalent years of experience
3-6 years in data science, computer science, software engineering, data engineering or related fields preferred
Proven expertise in designing and architecting advanced machine learning and AI solutions, including leading efforts to frame complex business problems, define scalable feature engineering strategies, and select optimal algorithms and architectures for enterprise-level applications.
Advanced proficiency in model training and optimization, with the ability to design efficient training pipelines, implement distributed training strategies, and apply sophisticated hyperparameter tuning techniques to maximize performance and scalability.
Deep experience in model validation and governance, including establishing rigorous evaluation frameworks, conducting comprehensive fairness and bias audits, and driving continuous improvement through advanced error analysis and benchmarking.
Proven expertise in production deployment of ML systems, including designing robust CI/CD pipelines, implementing containerization and orchestration (e.g., Docker, Kubernetes), and ensuring compliance with security and reliability standards across cloud environments.
Oversight of model monitoring and lifecycle management, including building automated monitoring systems, implementing drift detection and retraining workflows, and defining alerting mechanisms to maintain long-term model health and business impact.
Strong cloud expertise (AWS, Azure, GCP) for architecting scalable ML solutions, leveraging cloud-native services for data processing, model deployment, and monitoring at enterprise scale.
Expert-level programming skills in Python and SQL, with the ability to develop production-grade code, optimize queries for large-scale datasets, and mentor team members on best practices for coding and data management.

Nice To Haves

Master's Degree in Computer Science, Mathematics, Statistics, Economics or related field preferred

Responsibilities

Designing machine learning and AI models by framing business problems, engineering features, and selecting appropriate algorithms and architectures.
When designing solutions create processes that can be utilized for multiple business reasons and is adaptable.
Train models by preparing data, fitting algorithms, tuning hyperparameters, and validating robustness through cross-validation techniques.
Prior to development able to articulate the trade-off on different modeling techniques and implications when applied to business problem.
Validate model performance using statistical metrics, conduct fairness and bias assessments, and perform error analysis to refine model quality.
Create evaluation metrics and results that tie to business outcomes.
Deploy models into production environments by packaging them appropriately, integrating with systems, automating deployment workflows, including robust error handling and documenting for maintainability.
Monitor deployed models by tracking performance over time, detecting data drift, triggering retraining when necessary, and implementing logging and alerting mechanisms.
Communicate model results and trade-offs to leadership and stakeholders.