Assistant Manager, Fraud Analytics & ML Detection

Bupa Arabia; Jeddah; On-site; SAR 180,000 - 240,000; Full time

Job summary

Bupa Arabia seeks a data science professional to transform validated fraud analytics prototypes into robust detection models for healthcare. You will own model enhancement, feature engineering, calibration, validation, explainability, and continuous improvement across the lifecycle.

Collaborating with Actuarial, Claims Integrity, and FWA AI Engineering, you will ensure models stay effective against evolving fraud patterns after deployment and ready them for production in cloud environments.

Qualifications

• Bachelor's degree in a relevant field.
• Experience in health insurance, claims analytics, or financial crime analytics preferred.
• Strong Python/SQL/SAS skills with cloud analytics experience.

Responsibilities

• Model Industrialization Readiness: assess AI/ML prototypes and refine logic.
• Detection Model Development & Enhancement: retrain models and develop features.
• Model Validation & Performance Science: define metrics and validate outputs.
• Fraud Intelligence & Pattern Evolution: analyze outcomes and identify evolving fraud behaviours.
• Continuous Model Optimization: monitor deployed models and plan retraining.

Skills

Python

SQL

SAS

Machine Learning

Statistical Analysis

Feature Engineering

Model Validation

Fraud Analytics

Healthcare Analytics

NLP

GCP

Data Visualization

Cloud Analytics

Power BI

Stakeholder Management

Communication

Collaboration

Health Insurance

Education

Bachelor's Degree in CS/DS/AI or related field

Tools

SAS FWA

Power BI

Job description

Bupa Arabia seeks a data science professional to transform validated fraud analytics prototypes into robust detection models for healthcare. You will own model enhancement, feature engineering, calibration, validation, explainability, and continuous improvement across the lifecycle.

Collaborating with Actuarial, Claims Integrity, and FWA AI Engineering, you will ensure models stay effective against evolving fraud patterns after deployment and ready them for production in cloud environments.