Research Engineer, Model Evaluations

Design and implement evaluation methodologies that assess model capabilities across reasoning, safety, helpfulness, and harmlessness. Architect and scale the evaluation platform to adapt to rapidly advancing model capabilities. Build high-throughput evaluation pipelines that run during production training to provide real-time insights for training decisions. Analyze results to identify patterns, failure modes, and opportunities for model improvement, translating findings into actionable guidance. Collaborate with research teams to develop domain-specific evaluations and human-in-the-loop assessment approaches. Establish best practices for evaluation development, mentor teammates, and contribute to external publications and communications.