Director, Data Science and Machine Learning

POSITION TITLE: Director, Data Science and Machine Learning

Position Summary:

We’re looking for a data science leader to lead high-impact machine learning initiatives that drive business outcomes across the mortgage lifecycle—from acquisition to servicing. In this hands-on role, you’ll design and deploy machine learning models, advanced analytics, and experimentation frameworks that power data-driven decisioning at scale. You’ll translate complex business problems into scalable, production-ready data science solutions, with a focus on predictive modeling, customer segmentation, conversion optimization, and automation of decision workflows in the mortgage domain. This role reports to a senior leader in Data Science and is highly visible to top leadership, and across analytics, finance, and product teams.

Job Functions and Responsibilities:

To perform this job successfully, an individual must be able to perform each essential duty satisfactorily. 

• Design and implement machine learning models and statistical techniques across key mortgage workflows—such as risk scoring, churn prediction, segmentation, and structured document processing where applicable—to improve metrics like Conversion Rate, Delinquency Rate, and Customer Lifetime Value (CLV).
• Work closely with cross-functional partners (Product, Marketing, Engineering, and Finance) to identify opportunities for automation and insight generation using advanced analytics.
• Translate business problems across the mortgage lifecycle (e.g., underwriting, servicing, collections) into well-scoped modeling initiatives.
• Drive the development of automation-enhanced decision systems (e.g., pre-fill models for underwriting, early delinquency risk alerts, servicing escalation predictors) to enhance operational efficiency and user experience.
• Build robust data pipelines and modeling systems in collaboration with data engineering, ensuring scalability, monitoring, and model governance in production environments.
• Apply state-of-the-art techniques in self-supervised learning, graph-based modeling, and Bayesian mixture models to extract value from complex behavioral and relational data (e.g., referral networks, shared IP patterns).
• Contribute to a culture of excellence by publishing internal best practices, conducting peer reviews, and mentoring early-career data scientists informally (with no requirement to directly manage people).
• Actively contribute to the broader data science and machine learning community through publications, conference presentations, open-source contributions, or internal/external thought leadership, helping establish the company as a leader in applied AI for mortgage and financial services.
• Apply natural language processing (NLP) techniques as needed for tasks such as structured document parsing, entity extraction from disclosures, or classification of customer inquiries.
• Ensure compliance with data privacy regulations (e.g., GDPR, CCPA) and model auditability, particularly in financial and regulatory-sensitive applications.

Qualifications:

To perform this job successfully, an individual must have the following education and/or experience:

• Minimum education required: Masters or PhD in engineering/math/statistics/economics, or a related field
• Minimum years of experience required: 6 (or 3, post-PhD), ideally in mortgage, fintech, or financial services
• Required certifications: None
• Specific skill or ability needed 
  • Strong analytical and modeling skills, with experience applying machine learning to structured, unstructured, and semi-structured mortgage or financial data.
  • Demonstrated ability to translate business objectives into technical modeling goals and measurable success metrics. 
  • Experience building models for fraud detection, CLV estimation, or risk scoring is required; experience with NLP for structured document classification or disclosure review, compliance automation or underwriting is a plus.
  • Familiarity with modern ML Ops practices (data/model versioning, CI/CD, performance monitoring, and bias detection).
  • Experience with regulatory compliance, privacy, and auditability in model development (e.g., ECOA, RESPA, CFPB regulations).
  • Excellent communication skills for both technical and non-technical stakeholders, including ability to present complex results with cla