Senior Bayesian Risk Modeler

About Technosylva

Technosylva is a global leader in wildfire and extreme weather risk mitigation software. The Companys market-leading solutions, enhanced by AI and machine learning capabilities, provide real-time and predictive insights to support electric utility, insurance and government agency customers.

Technosylva has provided critical solutions for the past 26 years. In 2022 the organization entered a period of significant growth and transformation with investment from TA Associates, a leading growth PE firm, scaling to about 175 employees and offering its product in over 10 countries. In 2024 General Atlantic, a leading global growth investor, announced a strategic growth investment in Technosylva to support the company in its mission.

Role Overview

We are seeking an exceptional Senior Bayesian Risk Modeler with deep expertise in probabilistic modeling, Bayesian statistics, and Monte Carlo simulation to lead the evolution of our wildfire risk assessment platform. In this role, you will redesign and enhance our core annualized risk models, moving beyond deterministic approaches to comprehensive uncertainty quantification that supports critical utility decision-making during extreme weather events.

You will work at the intersection of advanced statistics, atmospheric science, and utility operationsdeveloping sophisticated probabilistic frameworks that accurately capture both the expected magnitude and uncertainty of wildfire risk. Your work will directly inform decisions affecting millions of utility customers and help prevent catastrophic wildfire events.

Main Responsibilities

• Redesign annualized risk frameworks using Monte Carlo simulation and Bayesian models to properly quantify uncertainty in wildfire risk estimates across weather scenarios.
• Develop and validate probabilistic models that propagate uncertainty from weather frequency, probability of failure (POF), probability of ignition (POI), and consequence estimates through to final risk outputs.
• Build calibrated prediction systems where stated confidence intervals accurately reflect true uncertainty.
• Optimize weather day selection algorithms using information-theoretic approaches to ensure representative coverage of risk scenarios.
• Apply extreme value theory to better characterize tail risks from rare but high-consequence weather events.
• Develop copula-based models to capture correlation structure in multi-dimensional weather exposure metrics.
• Maintain and run existing data pipelines that are used for risk calculation for our FireSight product.

Requirements

Education

• Ph.D. in Statistics, Biostatistics, Applied Mathematics, Computational Statistics, or related quantitative fields strongly preferred.
• A master's degree with exceptional depth in probabilistic modeling and 10 years of applied experience will be considered.
• Candidates with degrees in Physics, Econometrics, Operations Research, or Quantitative Finance with strong statistical backgrounds are encouraged to apply.

Professional Experience

• 6 years of experience in quantitative roles requiring sophisticated statistical modeling.
• Proven track record of deploying probabilistic models in production environments where uncertainty quantification directly informed critical decisions.
• Experience translating research-grade statistical methods into robust, scalable, and maintainable production systems.

Bayesian Statistical Modeling

• Proven mastery of Bayesian inference methodology, including hierarchical models, prior elicitation, posterior computation, and model selection.
• Hands-on experience implementing MCMC samplers (Gibbs, Metropolis-Hastings, Hamiltonian Monte Carlo) and understanding their convergence properties.

Monte Carlo Simulation & Uncertainty Quantification

• Extensive experience designing and implementing Monte Carlo frameworks for uncertainty propagation in complex systems.
• Expertise in rare event simulation and extreme value analysis for tail risk assessment.

Probability Theory & Statistical Foundations

• Working knowledge of copula methods for modeling dependence structures.
• Familiarity with extreme value distributions (GEV, Generalized Pareto) and their applications.

Programming & Implementation

• Advanced Python skills with demonstrated ability to write clean, efficient, well-documented code for complex statistical applications.
• Proficiency with probabilistic programming frameworks (PyMC, Stan, NumPyro, or equivalent).
• Comfortable working with large-scale geospatial and time-series datasets in formats like NetCDF and GeoTIFF.
• Experience with AI-assisted development tools (Cursor, Claude Code, or similar) for accelerating coding workflows.

Preferred

• Experience with wildfire risk modeling, catastrophe modeling, or climate risk assessment.
• Knowledge of utility infrastructure and power system operations.
• Experience with cloud computing technologies like Azure.