Tech Stack

Description

You will design and run machine learning experiments to understand and steer the behavior of powerful AI systems, contributing to AI safety research in areas like AI control and alignment stress-testing. Your work will involve building tooling, evaluating jailbreaks, and collaborating with teams to mitigate risks from advanced AI.

Requirements

• Significant software, ML, or research engineering experience
• Experience contributing to empirical AI research projects
• Familiarity with technical AI safety research
• Preference for fast-moving collaborative projects

Responsibilities

• Run multi-agent reinforcement learning experiments for AI safety
• Build tooling to evaluate effectiveness of LLM-generated jailbreaks
• Contribute to research papers, blog posts, and talks
• Test robustness of safety techniques by training models to subvert them
• Collaborate on safety-relevant projects with teams like Interpretability and Red Team