Senior ML Systems Engineer

Position: Senior ML Systems Engineer at Cohere
Location: Montreal

Elevate your career as a Senior ML Systems Engineer at Cohere in Montreal, Quebec. Focus on building sophisticated frameworks and tooling for large-scale language model training.

This mid-senior level role requires strong engineering experience in distributed training and HPC systems. You will design and maintain a comprehensive training framework, enhancing model reliability and performance. Engage with diverse teams and have autonomy in shaping impactful ML solutions across the organization.

Key Responsibilities:

• Build the training framework for large‑scale LLM training

• Design distributed training abstractions and memory management

• Enhance training throughput and stability on multi-node clusters

• Develop tooling for monitoring, debugging, and developer ergonomics

• Collaborate on Slurm setups and hardware configurations

Requirements:

• Experience in large-scale distributed training or HPC systems

• Familiarity with JAX, Slurm, or container orchestration

• Strong debugging skills in CUDA/NCCL and data pipelines

• Proven track record in ML developer tooling

• Excellent collaboration and judgment skills

Maximize your engineering skills in ML systems while making a significant impact at Cohere!