Member of Technical Staff - Post-training (reinforcement Learning & Generalization)

Member of Technical Staff at Cohere focused on post-training large language models using reinforcement learning. This role involves developing methods for generalization beyond training distribution, designing training recipes, environments, and evaluations, and diagnosing training dynamics.

Skills and focus areas

• reinforcement learning
• large language models
• out-of-distribution generalization
• reward design
• curricula
• regularization
• evaluation
• training dynamics
• research
• experimentation
• data infrastructure
• pre-training