Director of Predictive Analytics - Insurance

Orion180 Insurance Services•Melbourne, FL

5d•Onsite

About The Position

At Orion180, we believe the future of insurance belongs to organizations that can transform data into a deeper understanding of risk. Predictive and predictive modeling are not a support function—they are core strategic capability dictates how we grow, price risk, optimize our portfolio, and maintain a sustainable competitive advantage. As the Director of Predictive Analytics for property catastrophe insurance, you will lead our data science and modeling teams to measure, price, and manage risks from natural catastrophes and man-made hazards. In this role, you will shape Orion180's "view of risk" by overseeing the development of advanced AI-driven statistical and machine learning models that influence underwriting strategy, ratemaking, portfolio management, claims forecasting, and business planning. Positioned as a key leader within Enterprise Risk Management, you will drive initiatives that support our expansion into new markets and products. This role is both strategic and hands-on, offering a unique opportunity to join a growing organization where your work has direct executive visibility and measurable business impact.

Requirements

Master’s degree in Statistics, Data Science, Mathematics, Engineering, Actuarial Science, or a related quantitative field with 7+ years of relevant experience (PhD preferred).
Proven track record of developing, deploying, and monitoring predictive models within the property catastrophe insurance space.
Deep understanding of global reinsurance markets, regulatory environments, and commercial vendor models (e.g., RMS, AIR Worldwide, KatRisk).
Proficiency in Python, R, and SQL, along with machine learning frameworks (e.g., scikit-learn, Random Forests, XGBoost).
Advanced knowledge of pricing models, Generalized Linear Models (GLMs), and time-series forecasting for claims or weather-driven perils.
Background in GIS and geospatial modeling.
Experience building scalable data pipelines and managing models within cloud environments (AWS, Azure, or GCP) using MLOps tools (e.g., MLflow, SageMaker, Vertex AI).
Familiarity with insurance model governance, compliance, and documentation standards

Nice To Haves

PhD preferred

Responsibilities

Build and oversee machine learning models to predict the frequency and severity of property damage from natural disasters.
Establish and lead an analytics team to develop high-resolution supplementary analytical tools, underwriting solutions, pricing frameworks, and hazard maps.
Assist the exposure management and risk accumulation teams in validating and adjusting vendor and in-house models to develop a consistent view of climate risks, pricing, reserving, and capital models.
Translate complex analytical findings and modeling concepts into clear, actionable business strategies for executive leadership.
Partner closely with Underwriting, Actuarial, Claims, Finance, and Risk Management teams to ensure pricing and risk appetite align with corporate goals.