Training Infrastructure Software Engineer

Build scalable training software at Cohere, working remotely or from an office, with Python and distributed machine learning infrastructure. Support frontier model development across research and production.

As a Member of Technical Staff and Training Infrastructure Engineer, you will strengthen model training pipelines and ship state-of-the-art models. The work connects research with production through software engineering, experimentation, and infrastructure improvements. You will collaborate with researchers while contributing according to organizational needs.

Key Responsibilities:

• Design high-performance, scalable software for model training
• Improve training infrastructure and codebase performance
• Develop tools that accelerate training cycles
• Experiment on supercompute and data infrastructure
• Work with leading researchers in the field

Requirements:

• Extremely strong software engineering skills
• Proficiency in Python, JAX, PyTorch, and XLA/MLIR
• Experience with Kubernetes, Slurm, and Ray
• Experience using large-scale distributed training strategies
• Hands-on experience training large models at scale

Bring Python, distributed training, and infrastructure expertise to Cohere’s frontier model development across research and production.