AI Safety Specialist: Guardrails, Adversarial Testing & RLHF

Hyphen Connect; Warrnambool; Sur place; AUD 90 000 - 130 000; Plein temps

Résumé du poste

Hyphen Connect in Warrnambool, VIC, AU seeks an AI Safety Specialist to bolster the security and robustness of language models. You will conduct adversarial testing, implement guardrails, and guide RLHF alignment to ensure ethical alignment of AI behavior.

Responsibilities include testing LLMs and multimodal agents, deploying real-time filters for autonomous tool use, and developing constitutional AI principles to minimize risks in deployments.

Qualifications

• Background in cybersecurity, prompt engineering, or adversarial ML.
• Experience with jailbreak taxonomies and automated red-teaming frameworks.
• Strong analytical mindset for identifying edge cases.

Responsabilités

• Conduct adversarial testing on LLMs and multimodal agents.
• Implement guardrails and real-time filtering for autonomous tool use.
• Develop constitutional AI principles and assist with RLHF alignment pipelines.

Connaissances

Cybersecurity

Prompt engineering

Adversarial ML

Description du poste

Hyphen Connect in Warrnambool, VIC, AU seeks an AI Safety Specialist to bolster the security and robustness of language models. You will conduct adversarial testing, implement guardrails, and guide RLHF alignment to ensure ethical alignment of AI behavior.

Responsibilities include testing LLMs and multimodal agents, deploying real-time filters for autonomous tool use, and developing constitutional AI principles to minimize risks in deployments.