Principal Data Scientist

SUMMARY:
The Principal Data Scientist is a highly experienced individual contributor who serves as a technical authority in applying advanced analytics and machine learning to complex P&C insurance problems, including underwriting, pricing, and risk selection. This role owns the end‑to‑end analytical lifecycle, from problem formulation and model development through deployment, monitoring, and governance. Partners closely with Actuarial, MLOps, and IT to deliver scalable, production‑ready solutions. The Principal Data Scientist ensures long‑term model performance through rigorous validation, drift monitoring, and audit‑ready documentation, while advancing analytical best practices and evaluating emerging techniques relevant to commercial P&C insurance.
RESPONSIBILITIES/TASKS:
Acquires, organizes, and cleanses structured and unstructured data.
Conducts in-depth analysis to uncover trends, risks, and business opportunities.
Applies statistical modeling, machine learning, and advanced analytics to develop predictive and prescriptive solutions.
Evaluate solution performance using statistically rigorous methods and measure the impact to business outcomes.
Collaborate with MLOps and IT partners to transition solution prototypes from pilot validation into production environments.
Ensures ongoing model health through post‑deployment monitoring, drift detection, and audit‑compliant governance practices.
Creates and communicates results to senior level audiences of varying backgrounds, using business-facing presentations, reports, and dashboards.
Author and maintain comprehensive technical documentation for data lineage, codebases, results, and production changes.
Provides technical and project guidance, including peer review of work, for data science team.
Leads the evaluation of new analytic tools and processes.
Drivesinvestigation and adoption of advanced machine learning and AI innovations.
EDUCATION:
Bachelor’s Degree in Data Science, Statistics, Mathematics, Operations Research, Actuarial Science, Computer Science, Engineering, Physics or related technical field required. Advanced degree preferred.
EXPERIENCE:
10 years of experience in data science or related advanced analytics domains, including research and teaching, with 3+ years of technical leadership.
REQUIRED SKILLS/KNOWLEDGE/ABILITIES:
Broad experience supporting underwriting functions within multi-line commercial P&C insurance settings, including 3+ years of loss modeling for General Liability (aka Casualty) or Commercial Property.
Demonstrated expertise using Poisson, Gamma, and Tweedie distributions to build loss ratio, pure premium, and frequency–severity loss models for pricing.
Extensive experience leveraging supervised learning models (e.g., XGBoost, GLM, etc.) and unsupervised techniques (e.g., K-means clustering) to solve complex data science problems.
Advanced Python programming skills, including scikit-learn, and proficient ETL abilities using SQL.
Comfortable explaining machine learning models with partial dependence plots and SHAP values.
Ability to conduct experiments e.g., A/B Testing, to evaluate the causal impact of model-driven decisions.
Experience using version control tools such as Git and Azure DevOps.
Experience working in cloud computing environments such as Azure, AWS, GCP, etc.
Experience developing Agentic AI solutions to enable autonomous decision‑making and task orchestration.

PREFERRED SKILLS/KNOWLEDGE/ABILITIES:
In-depth understanding of Workers Compensation or Commercial Vehicle insurance.
Experience supporting Claims, Marketing, or Operations functions within P&C insurance settings.
Knowledge of actuarial concepts and terminology used in pricing and ratemaking.
Experience supporting both admitted and non-admitted commercial P&C lines.
Understanding of NLP concepts such as topic modeling, Word2Vec, sentiment analysis, OCR, etc.
Knowledge of advanced neural net architectures like LSTM, CNN, Transformers, Graph NN, etc.
Experience with causal modeling techniques such as Meta-learners, Causal Forest, Double ML, etc.
Experience programming in the R language.
Ability to build interactive dashboards using frameworks such as Plotly Dash, Power BI, Flask, etc.
Experience applying deep learning frameworks such as PyTorch, Tensorflow, Keras, etc.
ADDITIONAL INFORMATION:
The above statements are intended to describe the general nature and level of work being performed by people assigned to this classification. They are not intended to be construed as an exhaustive list of all responsibilities, duties and skills required of personnel so classified. This job description does not constitute a contract for employment.
PAY RANGE:
“Actual compensation decision relies on the consideration of internal equity, candidate’s skills and professional experience, geographic location, market and other potential factors. It is not standard practice for an offer to be at or near the top of the range, and therefore a reasonable estimate for this role is between $137,900 and $231,000.”
We are an Equal Opportunity Employer. We will not tolerate discrimination or harassment in any form. Candidates for the position stated above are hired on an "at will" basis. Nothing herein is intended to create a contract.

#AFG
Originally posted on Himalayas