Responsible Ai Intern

Centific Global

Seattle, WA, US
Base: $40 ph; bonus/equity: not specified; benefit...
Remote
Llm jailbreak attacks and defense
Agentic ai safety
Human-ai interaction vulnerabilities
Advance AI Safety by designing, implementing, and evaluating attack and defense strategies for LLM jailbreaks

Job Summary

  • Advance AI Safety by designing, implementing, and evaluating attack and defense strategies for LLM jailbreaks.
  • Evaluate AI Behavior by analyzing and simulating human-AI interaction patterns to uncover behavioral vulnerabilities.
  • Prototype workflows for multi-agent safety, including agent self-checks and regulatory compliance.

Matching Summary

Advance AI Safety by designing, implementing, and evaluating attack and defense strategies for LLM jailbreaks.

Salary

Base: $40 per hour; Bonus/Equity: Not specified; Benefits: Not specified

Skills & Requirements

Must-have

  • LLM jailbreak attacks and defense
  • agentic AI safety
  • human-AI interaction vulnerabilities
  • Python and PyTorch/JAX skills
  • design, implement, and evaluate attack and defense strategies

Nice-to-have

  • adversarial prompt engineering
  • multi-agent architectures
  • red-teaming and synthetic data
  • scalable training and deployment
  • public code artifacts and open-source impact

Key Requirements

  • Ph.D. student in CS/EE/ML/Security
  • Actively publishing in AI Safety, NLP robustness, or adversarial ML
  • Demonstrated research in LLM jailbreak attacks/defense, agentic AI safety, or human-AI interaction vulnerabilities
  • Proven ability to go from concept to code to experiment to result

Work Rights

Not specified

Tailored Resume

Cover Letter