Red Team Engineer, Safeguards at Anthropic

Role overview

Anthropic’s Safeguards team seeks a Red Team Engineer to find weaknesses in deployed AI products before attackers exploit them. The work spans product infrastructure, model behavior and abuse patterns such as coordinated account manipulation or payment fraud.

Responsibilities

• Run adversarial tests across product surfaces, combining multiple exploitation techniques and simulating realistic attack chains.
• Develop testing methods for agent systems, tool use and other emerging AI interactions.
• Build automated frameworks for continuous security and abuse assessment.
• Work with product, engineering and policy teams to convert findings into safeguards and detection metrics.

Qualifications

• Experience in penetration testing, red teaming or application security.
• Experience testing model jailbreaks and agent workflows for prompt-injection paths.
• Practical web security skills and tools such as Burp Suite, Metasploit or custom scripts.
• Experience creating security automation or LLM-specific test frameworks and communicating findings clearly.

Benefits

• The posting lists vacation and parental leave, flexible working hours and optional equity donation matching.
• Anthropic states that staff generally spend at least 25% of their time in an office; some roles require more. Check the posting for this role’s location and travel conditions.

About Anthropic

Anthropic develops AI systems with a focus on reliability, interpretability, steerability and safety.

Related jobs

• AI Enablement Engineer — Keyloop
• Embedded Linux Consultant Engineer (Remote/Anywhere) at Collabora — Collabora
• Enterprise Support Engineer at BlueCat Networks — BlueCat Networks
• AI Product Builder — Money Forward Vietnam
• Java Product Support Engineer at Sonatype — Sonatype
• Principal Engineer — Keyloop