About the role
The team trains large language models using reinforcement learning and develops methods for improved generalization. The engineer designs training recipes, environments, and evaluations to assess model learning and ensures successful research translates into production models.
Our summary of the posting; the original is on the Cohere careers page.
What the posting requires
- Required
- Research
- Reinforcement learning
- Experimental practice
- LLM
- Generalization
- Robustness
- Distribution shift
- PPO
- GRPO
- RLHF
- Preferred
- Distributed training
- JAX
- PyTorch
- FSDP
- Megatron
These tags are our reading of the posting. They can miss something; the original posting is the reference.