Researcher, Evals
Researcher, Evals
Researcher, Evals
Cartesia
Cartesia is a pioneering AI company focused on developing interactive intelligence that can process vast amounts of audio, video, and text data. The Evaluations Lead role involves designing frameworks to measure AI model capabilities, ensuring evaluations reflect real-world interactions and understanding. The position combines research, product development, and technical execution to shape the future of AI evaluation.
Qualification
- Experience designing or implementing evaluation frameworks for generative models
- Strong technical and analytical skills for translating research ideas into production systems
- Creativity in defining novel quantitative metrics for subjective qualities
- Excitement for building evaluation systems that connect research and real-world applications
- Curiosity and rigor in measuring meaningful progress in intelligent behavior
Responsibility
- Identify and define key model capabilities and behaviors for next-generation evaluations
- Develop and implement new evaluation pipelines with robust statistical analysis
- Partner with model training and research teams to integrate evaluation systems into development loops
- Prototype user studies and behavioral experiments for real-world evaluation grounding
- Design evaluation frameworks for generative models (audio, text, multimodal)
