Summary

Cohere is hiring a senior ml systems engineer to build and evolve the training framework for frontier-scale language models. The role focuses on distributed training abstractions, performance and stability on multi-node hpc clusters, and tooling for monitoring, logging, debugging, and reproducible runs, in a remote-friendly environment.

Responsibilities

• Build and own the llm training framework
• Design distributed training abstractions and strategies
• Improve training throughput and stability on clusters
• Develop tooling for monitoring, logging, and debugging
• Collaborate with infra teams on cluster and hardware support
• Investigate and resolve ml systems performance bottlenecks
• Build reproducible and debuggable large-scale run systems