Ai Safety Research Intern

Centific Global

Seattle, WA, US
Rate: $40 ph; bonus/equity: not specified; benefit...
Fully remote
Ph.d. student in cs/ee/ml/security
Python and pytorch/jax skills
Llm jailbreak attacks or defense experience
Join Centific to advance the frontiers of AI safety by designing attack and defense strategies for LLM jailbreaks and agentic workflows

Job Summary

  • Join Centific to advance the frontiers of AI safety by designing attack and defense strategies for LLM jailbreaks and agentic workflows.
  • You will own high-impact experiments from concept to prototype, directly contributing to robust security guarantees for scalable AI systems.
  • The role offers a competitive rate of $40 per hour while working alongside over 150 PhDs and data scientists on cutting-edge GenAI solutions.

Matching Summary

Join Centific to advance the frontiers of AI safety by designing attack and defense strategies for LLM jailbreaks and agentic workflows.

Salary

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

Skills & Requirements

Must-have

  • Ph.D. student in CS/EE/ML/Security
  • Python and PyTorch/JAX skills
  • LLM jailbreak attacks or defense experience
  • Agentic AI safety research background
  • Human-AI interaction vulnerability analysis

Nice-to-have

  • Experience with multi-agent architectures
  • Familiarity with red-teaming frameworks
  • Public code artifacts on GitHub
  • First-author publications in top venues
  • Knowledge of regulatory safety standards

Key Requirements

  • Active Ph.D. enrollment in relevant field
  • Publication record in AI Safety or NLP
  • Demonstrated research in adversarial ML
  • Ability to execute full research lifecycle

Work Rights

Not specified

Tailored Resume

Cover Letter