← All roles

Member of Technical Staff, Training Performance Engineer

Cohere· Modeling
Apply on company site
Hybrid (London), Hybrid (New York), Hybrid (Paris), Hybrid (Toronto), Hybrid (Montreal)CA$295k - CA$535kFull-timePosted 11mo ago

About the role

Performance-focused role within Cohere’s Pre-Training team, optimizing large language model training performance, throughput, and accelerator utilization through software engineering, low-level kernels, and profiling tooling.

Responsibilities

  • Design and write high-performant and scalable software for training.
  • Understand architectural modifications and design choices and their effects on training throughput and quality.
  • Write low-level CUDA, triton kernels to squeeze every last bit of performance from our accelerators.
  • Research, implement, and experiment with ideas on our supercompute and data infrastructure.
  • Learn from and work with the best researchers in the field.

Qualifications

  • Extremely strong software engineering skills.
  • Proficiency in Python and related ML frameworks such as JAX, Pytorch and XLA/MLIR.
  • Experience writing kernels for GPUs using CUDA, triton, etc
  • Experience using large-scale distributed training strategies.
  • Familiarity with autoregressive sequence models, such as Transformers.

Skills

Required:Software engineeringPythonGPU kernel developmentDistributed trainingTransformers / autoregressive models

Benefits

  • A weekly lunch stipend of $75/£75 or equivalent in local currency.
  • Full health and dental benefits, including a separate budget for mental health.
  • RRSP matching, 401K, Pension Scheme.
  • 100% Parental Leave top-up for up to 6 months, for either parent.
  • Annual enrichment benefits: arts & culture, fitness/wellness, quality time, workspace improvement credit; education & learning stipend.
  • 6 weeks of paid vacation (30 working days).
  • Budget for travelling to other offices if remote, plus an annual company offsite.
Last updated Oct 21, 2026. Interested? Apply on the company site.
Cohere logo

About Cohere

Support automation that's actually intelligent

Cohere resolves 40-70% of support requests using AI, over all channels, within just a few days.

artificial-intelligenceautomationliquid-ai-recruiting-target
Company profileWebsite

More open roles at Cohere