Staff Training Infra Engineer for Scalable ML Pipelines

WeHireYou; Paris (IN); Hybrid; USD 102,000 - 147,000; Full time

Benefits offered by this job

Lunch stipend

Health & dental benefits

Parental leave

Enrichment benefits

Vacation days

Travel between offices

Home office stipend

Job summary

Cohere is seeking a Member of Technical Staff, Training Infra Engineer to help design and scale our model training pipelines. You will contribute to production-grade training code, and bridge research with engineering across compute and data infrastructure.

You’ll work with researchers and engineers, using Python, JAX, PyTorch, Kubernetes, Slurm, Ray, and XLA/MLIR to push frontier models. Remote-friendly with office options in multiple cities, including Paris.

Qualifications

• Strong software engineering skills.
• Proficiency in Python and ML frameworks (JAX, PyTorch, XLA/MLIR).
• Experience with distributed training infra (Kubernetes, Slurm) and Ray.
• Experience training large models at scale and contributing to tooling.
• Bonus: publications at top venues (e.g., NeurIPS, ICML).

Responsibilities

• Design and write high-performant and scalable software for training.
• Improve training setup from an infrastructure and codebase performance standpoint.
• Craft and implement tools to speed up training cycles and improve infrastructure.
• Research, implement, and experiment with ideas on supercompute and data infrastructure.
• Learn from and work with leading researchers in the field.

Skills

Python

ML frameworks

Distributed training

Software engineering

Research collaboration

Tools

Kubernetes

Slurm

Ray

JAX

PyTorch

XLA/MLIR

Job description

Cohere is seeking a Member of Technical Staff, Training Infra Engineer to help design and scale our model training pipelines. You will contribute to production-grade training code, and bridge research with engineering across compute and data infrastructure.

You’ll work with researchers and engineers, using Python, JAX, PyTorch, Kubernetes, Slurm, Ray, and XLA/MLIR to push frontier models. Remote-friendly with office options in multiple cities, including Paris.