Summary

Cohere is hiring a member of technical staff to advance model post-training and bridge research to production. The role involves designing and writing scalable training software, post-training models to sota performance, and experimenting with techniques for sft and rl regimes on cohere’s compute and data infrastructure. The company is remote-friendly with offices in multiple cities.

Responsibilities

• Design and write scalable training software
• Post-train models to reach sota performance
• Coordinate with specialist teams to improve models
• Craft and implement techniques for sft and rl
• Research, implement, and experiment on infrastructure
• Support research by writing production code