Frontier AI Red Team Specialist

Obsidian; New York (NY); Remote; USD 90,000 - 130,000; Full time

Job summary

Cincinnatus LLC is seeking specialists to red-team frontier AI models, crafting multi-step tasks and benchmarks to reveal vulnerabilities and edge cases. You will work in a collaborative, research-focused environment, designing, testing, and documenting high-quality evaluation tasks for leading AI systems.

The role supports remote, full-time engagement within the United States, with a flexible weekly cadence around 35 hours and close coordination with researchers to iteratively improve

Qualifications

• MSc or PhD in a STEM field or equivalent research experience.
• 1+ years in research, research-engineering, security, or AI evaluation.
• Ability to identify vulnerabilities or failure modes in LLMs/ML systems via red teaming or evaluation.
• Proficiency in Python and Git for scripting probes and analyses.
• Strong knowledge of LLM capabilities and evaluation techniques.
• Experience in AI training, model evaluation, or benchmark authoring preferred.
• Attention to detail and independent problem solving with clear written communication.
• Available to commit ~35 hours per week.

Responsibilities

• Probe frontier AI models to identify how they behave on coding, ML, and analysis tasks and where they quietly fail.
• Design challenging tasks that are hard for models but fair to grade.
• Document findings with clear evidence and reproducible steps.
• Strengthen tasks by closing loopholes, shortcuts, and grading gaps in collaboration with authors.
• Share insights with researchers and experts to improve benchmarks.

Skills

Python

Git

LLM evaluation

Research analysis

Adversarial testing

Written communication

Education

MSc/PhD in STEM

Job description

Cincinnatus LLC is seeking specialists to red-team frontier AI models, crafting multi-step tasks and benchmarks to reveal vulnerabilities and edge cases. You will work in a collaborative, research-focused environment, designing, testing, and documenting high-quality evaluation tasks for leading AI systems.

The role supports remote, full-time engagement within the United States, with a flexible weekly cadence around 35 hours and close coordination with researchers to iteratively improve