Find your next role
Discover roles across ambitious teams
13,720 jobs
- More jobs
micro1
AI Jailbreak & Prompt-Injection Security Expert
Remote · Mid level
$50–$90 / hour
About the role
Contribute to AI safety by designing and implementing methodologies to evaluate AI system security. Help train next-generation AI systems through real-world input and adversarial testing.
What you’ll do
• Design methodologies for AI safety and ethical jailbreaks. • Create strategies to uncover adversarial bypass patterns. • Develop regression test suites for vulnerabilities. • Build evaluation frameworks against adversarial threats. • Collaborate with stakeholders to improve model safety.
What you’ll bring
• 2+ years in adversarial ML or AI security. • Experience with ethical jailbreaks and prompt injection. • Advanced degree or equivalent experience. • Recognized in AI security community. • Strong communication and documentation skills.