THE OPPORTUNITY
A role worth
a closer look.
Shape the measurements used to judge Claude during production post-training. Monitor live evaluation runs, investigate changes in model behaviour and improve the dashboards that guide training and release decisions. Partner across research teams to make evaluation results timely, interpretable and trustworthy under operational pressure.
Experience highlighted by the employer
- Experience designing and analysing ML evaluations at scale, with strong production Python skills.
- Ability to investigate ambiguous results and communicate rigorous recommendations across research teams.
Location & eligibility
Based in San Francisco or Seattle. Anthropic states at least 25% office attendance, with some roles requiring more. Visa support is considered but is not guaranteed for every candidate.
Compensation
USD $500,000–$850,000 / year. Employer-published annual salary range. Statistics and experiment-design experience are preferred qualifications.
Before you apply
This is TaskBull’s concise summary, not the employer’s full job description. Review the complete responsibilities, terms, accessibility information and privacy notice on the original application page.
Read the full employer listing ↗