Finance AI Model Auditor & Evaluator

Planet Pharma; San Francisco (CA); On-site; USD 140,000 - 190,000; Full time

Job summary

Planet Pharma is seeking an AI Trainer – Finance in San Francisco to craft realistic finance tasks and evaluate frontier AI outputs. You will handle valuation and modeling, credit analysis, investment memos, budgeting, and treasury/risk considerations, using real-world files and self-directed work. The role emphasizes clear written reasoning and model judgment.

The work is remote-friendly? No, on-site in San Francisco, with long-form tasks and flexible scope.

Qualifications

• 2+ years of hands on experience in corporate or investment finance.
• In progress Bachelor’s degree or higher in finance, economics, business or a quantitative field.
• CFA, FRM or securities licenses welcome but not required.
• Builds and audits models: three-statement, DCF, comps, credit or budget models.

Responsibilities

• Design challenging, realistic finance tasks drawn from your day-to-day work.
• Run tasks through frontier AI models and evaluate deliverables against professional standards.
• Compare two model outputs, document performance and gaps.
• Write detailed grading rubrics and explain why responses pass or fail.
• Flag concrete failures with clear evidence (logic, assumptions, sources).
• Contribute across valuation, credit, FP&A, markets, treasury and risk; refine tasks.

Skills

Valuation modeling

Credit analysis

Equity/Fixed income research

Portfolio management

Corporate finance & FP&A

Treasury & liquidity

Model auditing

Education

Bachelor’s degree in finance/economics/business or quantitative field

Tools

Google Workspace

Excel

Job description

Planet Pharma is seeking an AI Trainer – Finance in San Francisco to craft realistic finance tasks and evaluate frontier AI outputs. You will handle valuation and modeling, credit analysis, investment memos, budgeting, and treasury/risk considerations, using real-world files and self-directed work. The role emphasizes clear written reasoning and model judgment.

The work is remote-friendly? No, on-site in San Francisco, with long-form tasks and flexible scope.