Jobs

Callosum is seeking a Research Engineer, Evals (Member of Technical Staff) to lead the development of rigorous evaluation methodologies for agentic AI systems. This role involves designing benchmarks for complex agentic behaviors, building robust statistical methods, and transforming raw traces into structured evidence to localize failures. The ideal candidate will have a strong research background, deep hands-on experience with LLMs in agentic settings, and excellent engineering and communication skills. You will play a critical role in shaping the company's future by providing independent, evidence-based insights into AI system performance.