General Counsel for AI Legal Model Evaluation

Provide senior legal subject-matter expertise to a GenAI research team, creating authoritative instruction specifications, golden solutions, and evaluation benchmarks so frontier AI models reason correctly about real legal work. This role centers on hands-on legal judgment, translating professional standards into explicit tests and review criteria that guide model improvement.

• Review and quality-assure legal knowledge tasks and model outputs, identifying missing behaviors, superficial reasoning, and answers that appear plausible but would not pass professional scrutiny.
• Author high-quality instruction specifications and produce golden solutions that define correct legal work for model training and evaluation.
• Design challenging legal tasks and benchmarks, and help build legal-specific evaluation sets and tools together with research teams.
• Calibrate standards with researchers and adjacent specialists, converting tacit professional judgment into explicit, teachable criteria and consistent evaluation rules.
• Work inside the clients tooling, using client-issued accounts and equipment alongside research teams to implement and test legal tasks and evaluations.

• Juris Doctor, from an accredited law school, highly ranked schools preferred.
• 5 or more years of substantive post-qualification legal practice at a reputable institution, for example an established law firm, corporate legal department, regulatory body, or court. Internships and clerkships alone do not satisfy this requirement.
• Deep specialization in at least one substantive practice area, such as corporate and transactional, litigation and dispute resolution, regulatory and compliance, intellectual property, employment and labor, or tax.
• Demonstrated seniority and progression, such as partner, of counsel, counsel, senior associate, senior in-house counsel, or general counsel, with real ownership of matters.
• Active 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 errors.
• Excellent written communication, including the ability to provide precise, well-structured written feedback.
• Ability to commit reliably to 40 hours per week for the initial engagement period.

• Employment type: Full-time, W-2 employee of Cincinnatus LLC, placed to work as part of the clients extended workforce.
• Initial engagement length: 6 months, with the expectation of a 40 hours per week commitment.
• Work environment: Hybrid, Bay Area, California. You must live in the Bay Area and be able to work on-site with the client''s team multiple days per week when required. This is not a remote role.
• Relocation: Candidates not already based in the Bay Area must be willing to relocate at their own cost before the engagement starts. Relocation assistance is not provided.
• Provisioning: You will be provided with client-issued accounts and equipment, and will work inside the clients tools alongside their research teams.
• Position type: Structured, role-based employment integrated into enterprise workflows, not a freelance or purely project-based engagement.

• Hourly rate: $100 to $150 per hour.
• Paid via W-2 payroll administered by Cincinnatus LLC.

• Must be legally authorized to work in the United States and able to be employed as a W-2 employee of Cincinnatus LLC.
• Active bar admission to at least one U.S. state, in good standing, is required.
• Must reside in, or be willing to relocate to, the Bay Area, California, at candidates expense if not already local.

Submit your application through the hiring channel where you found this posting. Selected candidates will become W-2 employees of Cincinnatus LLC, complete onboarding and payroll setup with Cincinnatus, and then be placed to work on-site with the clients research team using client-provisioned accounts and equipment.

Cincinnatus LLC is an equal opportunity employer and provides reasonable accommodations for qualified applicants with disabilities. Employment decisions are made without regard to legally protected characteristics.