💼 About This Role

You'll join Safety Systems to identify and mitigate future AGI misalignment risks by designing worst-case demonstrations and adversarial evaluations. Your work will directly influence product safety and lower the chance of catastrophic outcomes. You'll collaborate with world-class researchers and have access to cutting-edge models.

🎯 What You'll Do

• Design worst-case demonstrations of AGI alignment risks
• Develop adversarial and system-level safety evaluations
• Build automated red-teaming infrastructure and tools
• Conduct research on alignment technique failure modes

📋 Requirements

• 4+ years experience in AI red-teaming, security research, or adversarial ML
• Strong research track record with publications or high-impact internal work
• Fluency in modern ML/AI techniques and large-scale codebases
• Ability to communicate complex findings to technical and non-technical audiences

✨ Nice to Have

• PhD in computer science, machine learning, security, or related field
• Experience with automated red-teaming and evaluation infrastructure
• Cross-functional collaboration skills spanning research, engineering, and policy

🎁 Benefits & Perks

• 🚀 Shape safety practices at the AGI frontier
• 🔬 Access to cutting-edge models, tooling, and compute
• 🤝 Collaborative, mission-driven environment
• 💰 Competitive compensation, equity, and benefits