Senior Counsel for AI Model Evaluation

Play a central role on a GenAI research team by applying hands-on legal practice experience to improve how frontier AI models perform real legal work. In this position you will evaluate model outputs, create high-quality instruction specifications and authoritative solutions, and design benchmarks that measure meaningful improvement. This role is focused on practicing legal specialists at the counsel or senior associate level who can translate real-world legal judgment into testable, teachable criteria.

• Data quality assurance and review, vetting legal knowledge work tasks and model outputs for missing behaviors, superficial reasoning, and answers that appear well written but would not withstand professional scrutiny.
• Create instruction specifications and produce golden solutions that define what a correct legal answer looks like in practice.
• Design challenging legal tasks and evaluation sets to deepen domain capabilities and build legal-specific benchmarks alongside research teams.
• Calibrate standards with researchers and adjacent specialists, making tacit legal judgment explicit and consistent.

• Juris Doctor degree from an accredited law school, strongly preferred from a highly ranked institution.
• 8 to 15 years of substantive post-qualification legal practice at a reputable law firm, corporate legal department, regulatory body, or court. Internships and clerkships alone do not satisfy this requirement.
• Current or prior seniority at the counsel, senior associate, senior in-house counsel, or Assistant General Counsel level, with real ownership of matters.
• Deep specialization in at least one substantive practice area, for example corporate and transactional, litigation and dispute resolution, regulatory and compliance, intellectual property, employment and labor, or tax.
• Admission to at least one U.S. state bar, in good standing.
• Practical experience using large language models in professional work, with the judgment to distinguish well-reasoned answers from plausible-sounding but incorrect responses.
• Excellent written communication, including the ability to provide precise, well-structured written feedback.
• Ability to commit reliably to 40 hours per week for an initial engagement of 6 months.

• Employment type: Full-time, W-2 employment with Cincinnatus LLC, with placement at a leading AI research lab as part of their extended workforce.
• This is a hybrid role based in the Bay Area, California. You must live in the Bay Area and be available to work on-site with the client multiple days each week when required. This is not a remote role.
• Candidates not already based in the Bay Area must be willing to relocate at their own expense; relocation assistance is not provided.
• You will be provisioned with client-issued accounts and equipment, and will work inside the client''s tools alongside their research teams.
• Roles filled through Cincinnatus are structured, role-based positions rather than project-based freelance engagements, and typically involve close collaboration with a client''s internal teams and integration into enterprise workflows.

Hourly pay range: 85 to 120 per hour.

• Must be authorized to work as a W-2 employee in the United States.
• Must reside in the Bay Area and be able to work on-site as described above.

Cincinnatus LLC will serve as the employer of record for this position, administering employment, onboarding, payroll, benefits, and compliance. If selected, you will be hired by Cincinnatus and placed to work directly with the client research team under the terms described above.

If you are earlier in your career, or if you are at partner or General Counsel level, please consider the corresponding Legal Domain Expert listings that match those seniority levels and rates.

Cincinnatus provides equal employment opportunities and does not discriminate on the basis of any protected characteristic. Reasonable accommodations are available for qualified applicants with disabilities during the application process.