Research Scientist/Engineer (Agentic Systems)

We are seeking a Research Scientist to build autonomous, large-scale environments that stress-test Large Language Model (LLM) agents to failure modes, thereby studying their breakdown mechanisms. You will work at the intersection of AI safety and software engineering to create realistic multi-agent scenarios, run end-to-end experiments against frontier models, and catalog concrete failure behaviors. This role involves converting research findings into internal models of agent behavior and public publications to help build robust AI guardrails at scale.