💼 About This Role

You'll investigate how adversaries exploit generative AI systems, including LLMs and autonomous agents. You'll translate research into defensive capabilities for Origin's endpoint security product. This role combines deep technical curiosity with hands-on experimentation and real-world impact.

🎯 What You'll Do

• Research adversarial misuse of LLMs and autonomous agents
• Develop and test jailbreak and prompt injection techniques
• Implement research findings into product improvements
• Produce public security research content and talks

📋 Requirements

• 5+ years in offensive or defensive security research
• Ability to code for research tooling development
• Explain complex concepts to executives and technical teams
• Adaptable to fast-paced startup environment

✨ Nice to Have

• Experience exploiting generative AI systems and jailbreaks
• Familiarity with RL, model interpretability, or safety research
• Proficiency in Rust, C/C++, or operating system internals

🎁 Benefits & Perks

• 🏖️ Flexible PTO
• 🏥 Generous healthcare
• 🏠 Home-office support
• 📈 Equity in high-growth startup

🚩 Heads Up

• Requirements list 5+ years but role title Senior suggests 6-9 years typical
• No explicit salary range provided in listing