AVP, Data Science & AI

Description and Requirements

Position Summary

The AVP, Data Science & AI , leads the delivery of data science and AI solutions for the Japan business and reports to the Japan CDO . The role is accountable for the full solution lifecycle, from problem definition and data sourcing through production deployment, monitoring, retraining or retirement, and measurement of realised business value.

This is a hands-on leadership role that combines strategic contribution, business partnership and technical depth. The role holder remains directly involved in solution design and code review, leads the adoption of modern data science, engineering and AI tooling, has first-line accountability for the quality of data used by the team, and leads the responsible use of AI in the solutions delivered.

Business requirements are often unclear at the outset: The role holder is expected to work with business stakeholders to define the decision or problem, agree scope and success measures, and build trust through transparent, consistent delivery.

Key Responsibilities

1. Data Science and AI Delivery

• Lead the end-to-end lifecycle of analytical and AI products, including problem framing, data sourcing, feature development, modelling, validation, deployment, monitoring, retraining and retirement.
• Build and prioritise the Japan use-case portfolio with the Japan CDO and business sponsors. Priority areas include underwriting triage, claims fraud and leakage detection, call center analytics and operational efficiency, agent and distribution productivity, customer acquisition and lead scoring, next best action, cross-sell.
• Establish and maintain technical standards for model development, validation, documentation and monitoring.
• Lead the application of generative AI and large language model solutions where appropriate, including Japanese-language document processing, agent and adviser support tools, call center knowledge assistants.
• Lead the selection and application of small language models (SLMs) for focused business use cases where privacy, latency, cost or deployment constraints make them appropriate, including task-specific evaluation, adaptation, retrieval, guardrails and production monitoring.
• Monitor deployed models for drift and performance degradation, and retrain or retire them as required.
• Maintain hands-on technical involvement, including design and code review, and explain and defend the team's models to Internal Audit, Risk and supervisory stakeholders.

2. Engineering and Productionisation

• Own the path from experiment to production, including deployment patterns, model registry, feature reuse, continuous integration and delivery, testing, observability and cost management.
• Promote disciplined, secure software engineering practices using Git-based collaboration and automation platforms, such as GitHub or comparable tools, including version control, pull requests, code review, testing and continuous integration and delivery.
• Own the curated data products, reusable features and analytics engineering discipline on which the team depends, ensuring consistency across models, reporting and agreed business definitions.
• Enable governed self-service and conversational analytics using fit-for-purpose capabilities, such as Databricks Genie or comparable tools, with trusted data, consistent metrics, business context and quality controls.
• Partner with the local and global platform teams on the underlying data platform, and contribute Japan requirements to the platform roadmap.

3. Data Governance and Data Quality

• Support on first-line data quality for all data used by the team, including quality rules embedded in pipelines, automated issue detection, root-cause analysis to source, and remediation agreed with system owners.
• Support on our data governance framework and assist our data stewards on lineage, metadata, retention and access.

4. Responsible Use of AI and Privacy

• Lead first-line implementation of the Responsible Use of AI framework in Japan, covering fairness and bias testing, decision-appropriate explainability, human oversight design, documentation and traceability.
• Apply heightened controls to customer-impacting decisions, including underwriting acceptance and rating, claims decisioning and fraud referral.
• Ensure compliance with the Act on the Protection of Personal Information, including requirements related to special care-required personal information, purpose limitation and cross-border transfer to regional and global headquarters.

5. Business Partnership and Global Alignment

• Act as the primary Data and Analytics counterpart for designated business functions.
• Translate unclear business requirements into scoped problems with a defined decision, success measure and accountable owner.
• Measure and report realised business value together with the business sponsor.
• Conduct workshops and training to strengthen business understanding of analytics and artificial intelligence.
• Adopt Enterprise frameworks, platforms and reusable assets where suitable, and represent Japan-specific requirements in regional and global forums.

6. Team Leadership

• Lead, develop and retain a team of data scientists, machine learning engineers, analytics engineers and analysts, together with external partners and vendors.
• Manage hiring, onboarding, performance and technical career development.
• Establish working practices including design review, code review and post-incident review.
• Manage team capacity across competing business demands.
• Maintain close and regular engagement with team members.

Required Qualifications

• Approximately 10 years or more of experience in data science, machine learning or advanced analytics, including at least 3 years leading technical teams.
• Demonstrated record of models deployed to production and in active business use.
• Experience with model governance, model risk management or responsible artificial intelligence in a supervised environment, including pre