VPII, Data Science Solutions

Job Overview:

The Vice President II, Data Science Solutions will lead the applied intelligence function within AI Business Solutions (ABS). This leader focuses on applied intelligence and decision systems, rapid prototypes, and evaluation standards, driving adoption across the company's product and platform teams. 

Reporting to the SVP of AI Business Solutions, this role partners closely with Technology and Governance to deliver intelligence that is measurable, explainable, and ready for regulated deployment. The role leads a high-output team that moves quickly from hypothesis to validated proof, then works with engineering to scale what works. 

This hybrid role must sit out of our Fort Mill SC or NYC office at least 3 days a week.

Responsibilities:

• Enterprise Intelligence Capabilities: Build and validate models and patterns that help applications make better decisions in the moment. This includes classification, entity extraction, prioritization, recommendation, anomaly detection, propensity and risk scoring, and other predictive capabilities used across advisor, assistant, investor, and operations experiences. 
• Event Mesh Development: Partner with engineering and governance to build the Event Mesh, the enterprise trigger and intent capability that converts messy signals into normalized events. Define the event vocabulary, intent and entity schemas, scoring approaches, and the measurement hooks required for downstream applications to subscribe and act reliably. 
• Document Fabric: Build, rent, or buy document intelligence capabilities that turn unstructured documents into deterministic, structured outputs. Focus on high-value document moments such as trust documents, forms, signatures, and other common workflow blockers. Ensure outputs are traceable and usable in production systems. 
• Decisioning and Recommendations: Develop models that shape the advisor, assistant, and operations day by prioritizing work, recommending next steps, and reducing avoidable rework. Partner with product leaders to define where recommendations belong in the workflow and how adoption and outcomes will be measured. 
• Rapid Prototyping and Validation: Operate a fast-cycle experimentation function that moves from hypothesis to validated proof in weeks. Build prototypes, measure impact, document results, and produce production-ready specifications for Technology to implement at scale. 
• Evidence Engine and Evaluation Standards: Establish rigorous evaluation methods for ML and LLM-enabled capabilities, including offline evaluation, online experiments where feasible, scenario-based testing, and monitoring signals. Build reusable scorecards and evidence packs that meet governance expectations and reduce friction in reviews. 
• Outcome Spine and Measurement: Partner with Business Reporting and Analytics and Technology to define the standard outcome instrumentation for intelligence-backed features. Capture events that show what was suggested, what was used, what was overridden, and the resulting outcome, so models improve over time, and value can be measured. 
• Label Foundry and Data Flywheel: Build the governed labeling and adjudication approach that turns operational outcomes into durable training data. Use targeted labeling and active learning to focus effort on the cases that change decisions and reduce cost. Grow proprietary labeled datasets tied to real outcomes, governed for privacy and retention. 
• Model Catalog and Reuse: Maintain a discoverable library of validated models, scorecards, and usage guidance so teams can adopt proven capabilities without reinventing them. Track adoption, performance over time, and refresh cadence. 
• Policy Router and Tiered Behavior: Work with governance and risk partners to define standard behavior expectations for different levels of automation and decision support. Ensure each capability has clear confidence handling, safe fallbacks, and audit-ready logging. 
• Cross-Functional Partnership: Operate in a matrixed environment across Product, Technology, Business, and Operations and Risk, co-designing specifications and evaluation frameworks for engineering to implement at scale. Translate business problems into measurable hypotheses, and ship validated approaches. 
• Data Science Community of Practice: Drive standards, shared learning, and career development across data scientists in ABS and partner organizations. Raise technical and evaluation standards and prevent fragmentation. 

What are we looking for?

We’re looking for a builder‑leader who loves turning messy, high‑value problems into production‑grade ML and NLP systems - spanning classification, document understanding, recommendations, and modern transformer‑based approaches. This role is about owning the full lifecycle, from smart experimentation and evaluation to measurable business impact that shows up in real advisor, investor, and operations workflows. You’ll tackle document intelligence and GenAI in a regulated environment, designing systems that are not only powerful, but explainable, auditable, and trusted. Most importantly, you’ll lead and grow a high‑caliber data science team while shaping how AI actually gets used at scale - not as demos, but as durable capabilities that move the business forward.

Requirements:

• A Bachelor’s degree in Computer Science, Statistics, Machine Learning, or a related quantitative field. A Master’s or PhD is preferred, or equivalent experience. 
• 10 or more years of experience in data science or machine learning, with 5+ years leading teams delivering production capabilities. 
• Proven track record building and validating ML systems used in real products, not just research prototypes. 
• Proven experience with modern ML tooling and stacks (Python, PyTorch or TensorFlow, common transformer tooling, and cloud ML platforms). 
• Proven experience with retrieval-augmented generation (RAG), embedding-based retrieval, and vector stores to ground model outputs in enterprise knowledge, with attention to evaluation, freshness, and traceability. 

Preferences:

• Financial services or wealth management background, with familiarity across advisor and operations workflows. 
• Experience with document processing vendors or cloud document intelligence services. 
• Experience with event-driven architectures, intent classification, and shared capability layers used across multiple applications. 
• Experience modeling and leveraging entity and relationship-based intelligence (e.g., knowledge graphs or graph databases) to support reasoning, reuse, and cross-domain decisioning is a plus.