Lead Data Scientist - FS/Insurance

EIRE Systems (Hong Kong) Limited; Hong Kong; On-site; HKD 900,000 - 1,500,000; Full time

Job summary

EIRE Systems (Hong Kong) Limited is seeking a senior data scientist to drive customer lifecycle analytics across multiple product lines. You will build production-ready models, deploy and monitor them at scale, and collaborate with product, marketing, and retention teams to turn insights into actionable campaigns.

The role emphasizes scalable data pipelines in Databricks, model governance, and explainability with SHAP/LIME to ensure business impact and measurable KPIs across cross-sell and

Qualifications

• 8+ years in advanced data analytics such as customer segmentation, regression, and propensity modelling.
• Experience developing and productionising predictive ML models (Classification, Regression, Gradient Boosting, Survival Analysis).
• Hands-on with MLOps practices, orchestration, and model tracking tools (MLflow, Airflow, CI/CD).

Responsibilities

• Design, develop, and refine customer segmentation, propensity, future value, and risk-scoring frameworks.
• Build, evaluate, and refresh ML models to predict churn, lifetime value, and cross-sell opportunities.
• Implement end-to-end feature engineering and automated ETL/ML data pipelines in cloud environments (Databricks).
• Own model deployment, scheduling, orchestration, and ongoing performance monitoring.
• Translate model drivers into interpretable business narratives using SHAP/LIME for marketing and retention.
• Collaborate with product, marketing, CRM and tech teams to embed scoring in systems and campaigns.

Skills

Python

SQL

Databricks

PySpark

MLflow

Airflow

XGBoost

LightGBM

SHAP

LIME

MLOps

CI/CD

Tools

Databricks / PySpark

Job description

In this role, you will act as the predictive engine for customer lifecycle management across multiple product lines. You will bridge advanced machine learning with tangible business outcomes—building, productionising, and monitoring predictive models that drive cross-sell conversion, retention strategies, and churn mitigation.

This is a production-focused data science role where you will design scalable data pipelines in Databricks, deploy robust models, and collaborate directly with product, marketing, and retention teams to turn analytical insights into high-impact operational workflows.

Key responsibilities

• Customer Analytics & Scoring Frameworks: Design, develop, and refine customer segmentation, propensity, future value, and risk-scoring frameworks across multi-product portfolios using structured transactional, behavioral, and intermediary data.
• Predictive & Churn Modeling: Build, evaluate, and refresh machine learning models (including classification, gradient boosting, and survival models) to proactively predict customer churn, lifetime value, and cross-sell opportunities.
• Feature Engineering & Pipeline Automation: Perform end-to-end feature engineering and construct automated ETL/ML data pipelines in cloud environments (Databricks) to operationalise model outputs at scale.
• MLOps & Model Governance: Own model deployment, scheduling, orchestration, and continuous performance monitoring (handling data drift, feature decay, and model degradation in production).
• Explainability & Business Integration: Utilize model explainability techniques (e.g., SHAP) to translate complex model drivers into clear, interpretable business narratives and campaign targeting logic for marketing, retention, and executive stakeholders.
• Cross-Functional Collaboration: Partner closely with product management, marketing, CRM, and technology teams to integrate scoring outputs directly into operational systems and campaign tooling.

About you

At least 8 years of proven experience in advanced data analytics such as customer segmentation, regression, and propensity modelling

Proven track record developing and productionising predictive ML models (Classification, Regression, Gradient Boosting e.g., XGBoost/LightGBM, Survival Analysis).

Expertise in Python (pandas, scikit-learn), SQL, and Databricks / PySpark handling large multi-source datasets.

Hands-on experience with MLOps practices, orchestration, and model tracking tools (e.g., MLflow, Airflow, CI/CD).

Solid understanding of model drift detection frameworks and explainability methods (SHAP / LIME).

Demonstrated experience in customer analytics, CRM, or customer lifecycle/churn modeling.

Prior exposure to Insurance (Life & Health, General Insurance) or Financial Services is highly advantageous.

Proven ability to map machine learning outputs directly to measurable business KPIs (e.g., campaign conversion, retention rates).

Strong stakeholder management and ability to translate complex technical findings into non-technical business narratives.