Preference Model is an AI research company dedicated to building automated machine learning research engineering. Its primary focus is creating reinforcement learning environments that accurately reflect real-world complexity, featuring diverse tasks and robust reward functions for training AI models. The company's technical work spans reinforcement learning, machine learning research, cybersecurity, data infrastructure, and dataset creation.
The founding team brings expertise from organizations including Anthropic, Stripe, and Datology. Preference Model partners with leading AI labs in its pursuit of advancing automated ML research capabilities.
