Senior Data Scientist, Insurance Pricing

Porch; Seattle (WA); Remote; USD 139,000 - 194,000; Full time

Benefits offered by this job

Long-term incentive

Medical plan options

Dental plan options

Vision plan

401(k) plan with match

Flexible vacation

Company holidays

Job summary

Porch Group is seeking a Senior Data Scientist in Seattle with remote options to develop GLM-based pricing models and advanced ML solutions for homeowners insurance. You will lead modeling standards, mentor junior scientists, and collaborate with actuarial, product, and engineering teams to deploy models in production and ensure regulatory compliance.

You will explore non-traditional data sources, conduct A/B testing, and refine models for profitability and pricing accuracy, leveraging BigQuery

Qualifications

• 5+ years in data science with depth in pricing or risk modeling.
• Technical ownership delivering complex projects end-to-end and mentoring others.
• Experience building GLM-based pricing models in production for homeowners or similar lines.
• Proficiency in Python and SQL, with GLMs and gradient boosting models.
• Experience with experimental design, including A/B testing in high-traffic environments.
• Cloud-based data platforms (BigQuery, GCP) and MLOps practices.
• Familiarity with Confluence and Jira in Agile/Scrum settings.
• Nice to have actuarial background or insurance mathematics.

Responsibilities

• Establish modeling standards and best practices for insurance pricing and risk modeling.
• Provide code review, knowledge sharing, and guidance to junior data scientists.
• Build, validate, and deploy GLM-based models for homeowners pricing (frequency, severity, loss cost).
• Create ML models (GBMs, neural networks) for underwriting accuracy and profitability.
• Develop ensemble models for insured-level profitability, retention, and conversion, incl. LTV models.
• Incorporate non-traditional data sources (aerial imagery, satellite, govt records) for risk quantification.
• Collaborate with product, actuarial, engineering, and business stakeholders to integrate data science into workflows.
• Work with actuarial teams to file and monitor regulatory models.

Skills

Python

SQL

GLM-based modeling

Machine learning

A/B testing

BigQuery

GCP

Confluence

Jira

Statistics

Education

Masters or PhD in Statistics, Mathematics, Computer Science, or related quantitative field

Tools

scikit-learn

statsmodels

xgboost

lightgbm

Job description

Porch Group is building and deploying homeowners insurance pricing and risk models end-to-end, from early scoping through production use. In this Senior Data Scientist role, you will develop GLM-based pricing models, machine learning and ensemble profitability and retention models, and apply non-traditional geospatial data to improve localized risk estimation and pricing outcomes. The position is based in Seattle, WA with remote work available across eligible locations.

What you will do

• Help establish and apply modeling standards and best practices for insurance pricing and risk modeling across the team.
• Provide code review, knowledge sharing, and technical guidance to mentor junior data scientists.
• Build, validate, and deploy GLM-based models for homeowners insurance pricing, including frequency, severity, and loss cost.
• Create machine learning models such as gradient-boosted models (GBMs) and neural networks to support underwriting accuracy, competitive positioning, and profitability.
• Develop ensemble models to predict insured-level profitability, customer retention, and conversion, including customer lifetime value (LTV) models to prioritize marketing and underwriting strategies.
• Incorporate non-traditional data sources, including aerial imagery, satellite data, government records, and building permits, to quantify localized risk for pricing and underwriting decisions.
• Partner with product, actuarial, engineering, and business stakeholders to scope initiatives and integrate data science solutions into operational workflows.
• Work with actuarial teams to develop, file, implement, and monitor new predictive models that meet regulatory requirements.
• Apply rigorous deployment practices in high-traffic environments, including A/B testing, performance monitoring, and continuous refinement.
• Evaluate emerging approaches, including generative AI and LLMs, to identify potential competitive advantages.

What you bring

• 5+ years of experience in data science, with significant depth in insurance pricing or risk modeling.
• Proven technical ownership delivering complex projects end-to-end and supporting or mentoring less-experienced data scientists.
• Demonstrated experience building, validating, and deploying GLM-based pricing models in production, ideally for homeowners or other property/casualty lines, plus experience with machine learning models for non-pricing use cases.
• Proficiency in Python and SQL, including implementing GLMs and gradient boosting models using scikit-learn, statsmodels, xgboost, and lightgbm.
• Experience with experimental design, including building, deploying, and A/B testing models in high-traffic environments, as well as causal analysis.
• Experience with cloud-based data platforms (for example, BigQuery and GCP) and MLOps practices such as model training pipelines, versioning, and monitoring.
• Familiarity with Confluence and Jira for documentation and project tracking in an Agile/Scrum environment.
• Master’s or PhD in Statistics, Mathematics, Computer Science, or a related quantitative field (preferred).
• Nice to have: background in actuarial science or insurance mathematics, including collaborating with actuarial teams on regulatory model filing.
• Nice to have: experience with customer lifetime value, retention, or conversion modeling.
• Nice to have: geospatial or spatial data analysis experience (aerial imagery, satellite data, or property-level geographic datasets).
• Nice to have: exposure to generative AI and LLMs, including prompt engineering or fine-tuning for insurance or financial services use cases.
• Nice to have: exposure to experimentation frameworks and causal inference methods.
• Nice to have: experience with model governance, validation frameworks, and regulatory compliance in insurance.

Tools and technologies

• Python, SQL, scikit-learn, statsmodels, xgboost, lightgbm, GBMs, neural networks
• BigQuery, GCP, Confluence, Jira
• MLOps, A/B testing, generative AI, LLMs

Compensation and benefits

• Pay range: $138,800 - $194,300 per year
• Long-term incentive awards, subject to program guidelines and approvals
• Medical plan options (three (3) Medical plan options)
• Dental plan options (two (2) Dental plan options)
• Vision plan
• Critical Illness, Hospital Indemnity and Accident plans (voluntary basis)
• Pre-tax savings options including a partially employer funded Health Savings Account
• Employee Flexible Savings Accounts including healthcare, dependent care, and transportation savings options
• Company paid Basic Life and AD&D, Short and Long-Term Disability benefits
• Voluntary Life and AD&D plans
• Both traditional and Roth 401(k) plans with a discretionary employer match
• Supportlinc wellbeing program
• LifeBalance free resource for year-round discounts
• Flexible paid vacation
• Company-paid holidays of typically nine per year
• Paid sick time
• Paid parental leave
• Identity theft program
• Travel assistance
• Fitness and other discounts programs

Location and application details

• Location: Seattle, WA (remote)
• Workplace type: Remote
• Application window: anticipated to close in 2 weeks (10 business days) from 11/6/2026
• Remote positions not considered from: Alaska, Delaware, Hawaii, Mississippi, Nebraska, Montana, New Hampshire, West Virginia, District of Columbia