Quantitative Researcher

Qualified Intelligence insures the cost of fixing work done with AI. Companies now run real work on AI. Sometimes that work has to be fixed, and companies spend real money fixing it: engineering time, additional compute and model usage, outside help, data repair, and customer remediation. Existing policies weren't built for this category. There may be no lawsuit, no breach, and no third-party demand.

We are looking for a quantitative researcher to help make that risk measurable and underwritable. The work is applied research: define the exposure, study the data, build models, and test assumptions.

This is an early-career role for someone with strong quantitative skills, enough programming experience to analyze data, and curiosity about AI. You don't need insurance experience. You do need to be comfortable with probability, statistics, modeling, and clear thinking under uncertainty.

We work together in person in San Francisco. We offer full healthcare and benefits, and early hires will shape the product, the underwriting system, and the company culture. This is a small team, not a rotation program. You will work directly with the founders, underwriting, engineering, and outside insurance partners, and your work will inform how the company selects, prices, limits, or declines risk.

You might be a fit if. You have a strong academic record in mathematics, statistics, computer science, physics, engineering, operations research, economics, or a related quantitative field. You may be finishing an undergraduate degree, master's degree, PhD, or postdoc, or you may have recently graduated.

You enjoy problems where the data is messy and the right model isn't obvious.

You can go from idea to working analysis without waiting on someone else, whether that means writing code yourself or directing AI to write it. Python is the most useful language for this role, but the specific language matters less than your ability to work with data, build simulations, evaluate models, and explain your results.

You use AI tools regularly and they're part of how you work. Maybe you've built side projects with agents, used models to analyze data, written small tools for yourself, automated repetitive parts of your work, or tested what current systems can and can't do.

You have done independent work before: a thesis, research project, competition, open-source project, forecasting project, data science project, simulation, paper replication, agent experiment, or another project that shows you can work through an open-ended problem on your own.

The job. Build quantitative models for the cost of fixing work done with AI. Work with underwriting to define exposure variables, analyze early submissions and claims, estimate frequency and severity, study aggregation, test pricing logic, and build tools that help underwriters make consistent decisions.

You will spend time on data cleaning, exploratory analysis, simulations, model evaluation, and internal research memos. You'll build tools, run analyses, and write research that feeds directly into underwriting decisions.

What we look for. Strong probability and statistics. Strong programming. Good judgment about data. Clear written communication. Comfort working in a small team.

Finance or insurance experience is helpful but not required.

Apply. Email hiring@qualifiedintelligence.com with a resume or LinkedIn profile and a short note about a project or paper you are proud of. If you have built something with AI in your spare time, send that too.