Remote AI Security Analyst — Adversarial Safety & Risk

Alignerr Corp.; Madrid; Remote; EUR 83,000 - 152,000; Full time

Job summary

Alignerr is seeking AI Security Analysts to probe frontier AI models for vulnerabilities, focusing on adversarial conditions and safety alignment. This fully remote, hourly contract role offers flexible hours (10–40 hours/week) and work on some of the most technically interesting problems in AI security today.

You will analyze AI/LLM security scenarios, investigate prompt injection and data leakage risks, and document findings to improve model safety and reliability, with independent,

Qualifications

• Background in cybersecurity or related field.
• Familiar with threat modeling concepts applied to software systems.
• Analytically sharp with attacker mindset and defender documentation.
• Curious about edge cases, unexpected behaviors, and complex systems.
• Detail-oriented when evaluating risk, likelihood, and impact.

Responsibilities

• Analyze AI/LLM security scenarios under adversarial conditions.
• Investigate prompt injection, data leakage, model abuse, and system misuse.
• Classify security vulnerabilities and recommend mitigations.
• Evaluate AI system behavior and contribute structured findings.
• Work on high-stakes scenarios from frontier AI deployments.
• Complete task-based assignments independently on your own schedule.

Skills

Cybersecurity mindset

Threat modeling

Analytical thinking

Documentation

Independent work

Education

Bachelor's in Cybersecurity / CS

Tools

LLM security tools

Security assessments

Job description

Alignerr is seeking AI Security Analysts to probe frontier AI models for vulnerabilities, focusing on adversarial conditions and safety alignment. This fully remote, hourly contract role offers flexible hours (10–40 hours/week) and work on some of the most technically interesting problems in AI security today.

You will analyze AI/LLM security scenarios, investigate prompt injection and data leakage risks, and document findings to improve model safety and reliability, with independent,