Principal Data Scientist

Mariner Finance•Nottingham, MD

1d•$155,000 - $195,000•Remote

About The Position

Mariner Finance is hiring a Principal Data Scientist to serve as a principal-level technical leader and advanced data science contributor. The role develops, evaluates, scales, and improves high-impact AI, machine learning, and predictive modeling solutions across business areas such as credit risk, collections, loss mitigation, decisioning, pricing, fraud, and customer strategy. The position leads modeling initiatives end to end, from defining problems and conducting experiments through model development, production deployment, monitoring, and ongoing optimization. It applies advanced technical judgment to complex or ambiguous business problems and partners with business and technical stakeholders to improve portfolio performance, profitability, risk management, and decision-making. As an advanced technical resource within the Data Science function, the Principal Data Scientist provides technical direction, coaching, peer review, and escalation support. The role advances modeling practices, AI evaluation and measurement, reusable development patterns, and scalable approaches for applying AI responsibly. It works with cloud and data platforms, including AWS and Databricks, and collaborates with engineering, MLOps, technology, risk, and other business teams.

Requirements

Bachelor’s degree in Data Science, Statistics, Mathematics, Computer Science, Economics, or a related field; additional applicable experience may be substituted for a bachelor’s degree.
At least seven years of progressive data science, statistical modeling, or machine learning experience, including at least five years developing and operationalizing predictive or machine learning models.
Experience leading complex modeling initiatives through the full model lifecycle, from business problem definition and model development through production deployment, monitoring, and optimization.
Advanced proficiency in Python and SQL, with experience using commonly used machine learning, statistical modeling, and data science frameworks and libraries.
Advanced experience developing and deploying modeling solutions in cloud environments, including AWS.
Experience with modern data and analytics platforms such as Databricks or comparable technologies.
Advanced knowledge of predictive modeling, machine learning, artificial intelligence, statistical analysis, experimentation, model evaluation, performance measurement, and practical uses of AI in data science delivery.
Experience designing and evaluating statistically rigorous experiments and AI/ML evaluation approaches, translating model quality, reliability, drift, and business-impact results into actionable recommendations.
Experience developing, evaluating, and scaling production AI and machine learning solutions and partnering with engineering, MLOps, or technology teams to integrate models into operational systems.
Demonstrated ability to use AI-enabled tools or methods to improve the speed, quality, repeatability, or scalability of data science delivery.
Strong understanding of model lifecycle management, performance monitoring, documentation, and model risk considerations.
Experience developing or applying AI evaluation and measurement frameworks, including defining success criteria, testing failure modes, assessing production readiness, and communicating limitations and tradeoffs.
Ability to independently solve complex and ambiguous analytical problems and determine appropriate modeling approaches with limited direction.
Experience providing technical guidance, coaching, peer review, or mentorship to data scientists or technical team members.
Ability to serve as a senior technical thought partner to leadership by shaping modeling strategy, challenging assumptions, improving standards, and helping scale AI and machine learning capabilities.
Strong stakeholder management and communication skills, including the ability to explain technical concepts, risks, tradeoffs, and recommendations to technical and non-technical audiences.
Strong analytical, problem-solving, organizational, and decision-making skills.
Ability to manage multiple complex initiatives and priorities independently and collaboratively across functions.

Nice To Haves

Advanced degree in Data Science, Statistics, Mathematics, Computer Science, Economics, or a related quantitative discipline.
Experience in consumer lending, particularly with credit risk, collections, loss mitigation, decisioning, loss forecasting, optimization, or related modeling use cases.
Significant enterprise modeling experience in banking or financial services.
Hands-on experience with reinforcement learning, causal inference, multi-armed bandits, transformer-based models, optimization techniques, or similar advanced approaches, with sound judgment about when to use them instead of simpler traditional machine learning methods.
Experience working within established model risk management or model governance frameworks.

Responsibilities

Lead development, evaluation, and operationalization of complex predictive, machine learning, and AI models, including AI-enabled techniques and tools that improve model development, review, documentation, monitoring, optimization, and scale.
Own the technical lifecycle of assigned modeling solutions, including problem formulation, data assessment, feature engineering, model development, experimentation, validation, deployment, performance monitoring, recalibration, and enhancement.
Translate complex or ambiguous business challenges into analytical and modeling strategies; evaluate alternatives and recommend solutions based on business impact, feasibility, risk, and scalability.
Apply advanced statistical, machine learning, and artificial intelligence techniques, including predictive modeling, causal inference, optimization, reinforcement learning, multi-armed bandits, and transformer-based models where appropriate.
Design and lead rigorous experiments and evaluation approaches, including A/B testing, causal methods, AI/ML evaluation frameworks, and business-impact measurement.
Develop and operationalize scalable modeling solutions using cloud and data platforms such as AWS and Databricks, partnering with engineering, MLOps, and technology teams on production integration.
Monitor model performance and business outcomes; identify degradation, emerging risks, and improvement opportunities, and recommend recalibration, enhancement, replacement, or strategy changes.
Design and implement AI and machine learning evaluation frameworks measuring model quality, predictive performance, reliability, fairness where applicable, drift, stability, business impact, and fit for purpose.
Partner with Risk, Loss Mitigation, Technology, Data, and other stakeholders to understand priorities, shape technical approaches, communicate findings, and support adoption of model-driven recommendations.
Serve as a technical escalation point for complex modeling, experimentation, data, or implementation challenges.
Review and challenge modeling methodologies, technical designs, assumptions, and analytical conclusions to promote quality, consistency, scalability, and sound data science practices.
Provide technical guidance, coaching, peer review, and knowledge sharing to data scientists and other team members.
Advance modeling standards, reusable analytical approaches, methodologies, documentation practices, and technical patterns for the Data Science function.
Partner with senior leaders and stakeholders on model governance, documentation, validation, monitoring, controls, and lifecycle requirements.
Evaluate emerging modeling techniques, technologies, tools, and data sources and recommend opportunities to improve predictive performance, experimentation, operational efficiency, or business outcomes.
Conduct advanced portfolio analyses and recommend improvements related to portfolio performance, risk, and strategy.
Communicate analytical concepts, model performance, tradeoffs, risks, and recommendations to technical and non-technical audiences.
Perform additional functions depending on business needs.