Insurance and Actuarial Professional for AI Model Evaluation

Role Overview

Help advance frontier AI models by bringing senior insurance and actuarial judgment to the evaluation of real-world insurance work. You will work directly with an AI research and program management team, translating professional standards into tasks, solutions, benchmarks, and feedback that distinguish sound reasoning from plausible but incorrect answers.

Key Responsibilities

• Review insurance knowledge-work tasks and model outputs for missing behaviors, weak reasoning, flawed actuarial assumptions, misapplied policy language, and conclusions that would not withstand professional scrutiny.
• Write instruction specifications and golden solutions for insurance and actuarial problems.
• Create tasks that reflect real insurance practice, along with challenging benchmarks and evaluation sets.
• Help develop domain-specific skills and tools with the research team.
• Collaborate with researchers and adjacent-domain specialists to calibrate standards and convert tacit insurance judgment into clear, teachable criteria.

Qualifications

• Hold an FSA, ASA, FCAS, or ACAS credential, or an active state insurance license in adjusting, underwriting, or broking, such as an all-lines adjuster license.
• Actuarial candidates must have a degree in actuarial science, mathematics, statistics, or a related quantitative field.
• Have 4+ years of substantive experience with a carrier, reinsurer, broker, adjusting firm, or insurance regulator.
• Bring specialization in at least one area such as life and annuity, group life and disability, property and casualty, health, reinsurance, pricing and reserving, enterprise risk management, claims adjudication, or underwriting.
• Demonstrate senior-level progression, such as Actuary, Senior Actuary, AVP, Director, Chief Actuary, Claims Manager, or Head of Underwriting, with ownership of a book, reserve, or claims portfolio.
• Use large language models hands-on in professional work and can distinguish rigorous reasoning from a convincing but incorrect response.
• Have excellent written communication skills and can provide precise, well-structured feedback.

Work Terms

• Full-time W-2 employment, with placement on an extended workforce team supporting a leading AI lab.
• Commit reliably to 40 hours per week for an initial 6-month engagement.
• Hybrid role based in the Bay Area, California, requiring on-site work with the client team multiple days per week when needed.
• This is not a remote role. Candidates must already live in the Bay Area or relocate there at their own expense before the engagement begins. Relocation assistance is not available.
• Client-issued accounts and equipment will be provided, and work will be completed in the client’s tools alongside research teams.

Compensation

$60 to $100 per hour.

Eligibility and Application Process

Selected candidates are employed through an employer-of-record arrangement that administers W-2 employment, payroll, benefits, and compliance while the employee works within the client team. Equal employment opportunity is provided without discrimination on legally protected grounds, and reasonable accommodations are available to qualified individuals with disabilities and disabled veterans during the application process.