AI Security Analyst

The work

Find AI security weaknesses before they become operational problems. Design adversarial tests, assess model behavior and document actionable findings that help engineers strengthen system safety and reliability.

• Conduct red team exercises to uncover security flaws.
• Design and run adversarial prompts and unusual cases that probe model guardrails.
• Assess outputs for safe behavior, bias and compliance with policy.
• Document vulnerabilities, exploits and unexpected results in structured reports.
• Recommend practical mitigations with engineering colleagues.
• Track new AI threats, jailbreak techniques and changes in effective security practice.
• Help refine test protocols and assessment rubrics.

Experience and skills

• Solid cybersecurity foundations in threat modeling, penetration testing or vulnerability research.
• Hands on experience with AI or ML, language models or prompt engineering.
• Creative adversarial reasoning and the ability to identify overlooked weaknesses.
• Clear, actionable technical writing.
• Independent working ability within asynchronous task workflows.
• Familiarity with open source AI tools is helpful.
• Information security, ethical hacking or AI safety research experience is useful, but not mandatory.

Working arrangements

This hourly contract is remote, with an indicated commitment of 10 to 40 hours weekly and control over your schedule. The work provides experience at the intersection of cybersecurity and AI. Strong contributors may receive broader assignments and contract extensions.