Recidivism & Evasion Enforcement Specialist

Anthropic; Washington (District of Columbia); Hybrid; USD 245,000 - 285,000; Full time

Benefits offered by this job

Hybrid work arrangement

Job summary

Anthropic is seeking a Safeguards Enforcement Analyst for the account abuse team to build and run enforcement workflows that keep our products safe. The focus includes detecting recidivism and ensuring banned actors cannot evade a ban quickly.

You will identify evasion across identities, link accounts, and close re-registration paths while aligning with evolving AI policy practices. Strong SQL, data analysis, and collaboration are essential.

Qualifications

• Experience investigating ban evasion, multi‑accounting, or repeat fraud actors on adversarial platforms.
• Fluency in SQL and ability to analyze large account/event datasets.
• Experience with fraud or identity-linking signals and precision/recall tradeoffs.
• Rigor about evidence standards and the cost of false positives in severe-harm enforcement.
• A track record turning one‑off investigations into repeatable detection logic and policy.
• Strong written communication for briefs and recommendations for technical and non-technical stakeholders.
• Excellent judgment and collaboration with evolving priorities and workstreams.

Responsibilities

• Investigate evasion clusters end to end—from a single appeal or signal anomaly to the full linked actor network.
• Convert individual findings into durable systemic controls and detection proposals.
• Operationalize re-registration controls for high-severity ban populations.
• Partner with Engineering and Data Science on account-linking signals across identities.
• Build the recidivism measurement framework: return rates, speed of detection, and control impact.
• Author playbooks for contractor-supported evasion reviews with QA against gold standards.
• Keep up to date with emerging AI policy enforcement best practices and apply them to workflows.

Skills

SQL

Fraud analysis

Account linking

Evidence standards

Repeatable detection

Written briefs

Collaboration

Education

Bachelor's degree

Tools

Job description

Anthropic is seeking a Safeguards Enforcement Analyst for the account abuse team to build and run enforcement workflows that keep our products safe. The focus includes detecting recidivism and ensuring banned actors cannot evade a ban quickly.

You will identify evasion across identities, link accounts, and close re-registration paths while aligning with evolving AI policy practices. Strong SQL, data analysis, and collaboration are essential.