Performance Engineer, Scalable ML Training

Cohere; City Of London; Remote; GBP 110,000 - 190,000; Full time

Benefits offered by this job

Lunch stipend

Health and dental benefits

Pension/retirement plan

Parental leave top‑up

Learning stipend for conferences

Vacation days (6 weeks)

Travel to other offices

Home office stipend

Job summary

Cohere is seeking a Performance Engineer for the Pre-Training team to optimize our language models and systems. You will focus on improving training throughput and accelerator utilization, working at the intersection of software engineering, ML, and low-level kernel design.

We operate remotely with offices in London and other cities; candidates in ET to CET time zones are ideal for collaboration. Join a world‑class team driving efficient AI at scale.

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.

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.

Skills

Python

JAX

PyTorch

MLIR/XLA

Software engineering

Distributed training

Transformers

Tools

CUDA

Triton

Job description

Cohere is seeking a Performance Engineer for the Pre-Training team to optimize our language models and systems. You will focus on improving training throughput and accelerator utilization, working at the intersection of software engineering, ML, and low-level kernel design.

We operate remotely with offices in London and other cities; candidates in ET to CET time zones are ideal for collaboration. Join a world‑class team driving efficient AI at scale.