LLM Red Team Specialist – Remote Contract

About the role

We are seeking an experienced LLM Security Researcher to join our remote Red‑Team. You will design and execute adversarial conversations and task‑based scenarios that probe the limits of frontier large language models.

Key responsibilities

• Develop complex, multi‑turn adversarial dialogues and task scenarios aligned with project specifications.
• Write clear evaluation rubrics to measure model responses against defined behavioral targets.
• Iteratively test and increase difficulty of conversations on cutting‑edge LLMs.
• Deliver complete task packages, including transcripts, target behaviors, and rationale.
• Validate model outputs, documenting strengths, failure modes, and alignment with specifications.
• Maintain calibration with team leads and quality‑control contacts as requirements evolve.

Required profile

• Exceptional written English with clarity, precision, and strong organization.
• Prior experience in AI human‑data environments such as RLHF, SFT, evaluations, or prompt engineering.
• Deep familiarity with large language models and common failure patterns.
• Ability to work autonomously, interpret complex specifications, and execute them.
• Critical thinking and meticulous attention to detail.

Required skills

• Large language model evaluation
• RLHF (Reinforcement Learning from Human Feedback)
• SFT (Supervised Fine‑Tuning)
• Prompt engineering