LLM Post-Training Data Infrastructure Engineer

Cohere; Toronto; Remote; CAD 295,000 - 535,000; Full time

Benefits offered by this job

Lunch stipend

Health and dental benefits

RRSP matching / pension

100% Parental Leave top-up

Enrichment benefits

Education stipend

Paid vacation (6 weeks)

Office/travel and home office stipend

Job summary

Cohere is a leading security-first enterprise AI company building foundation AI models and end-to-end products. We are hiring for a role that builds and owns pipelines for post-training data used to train large language models.

You will collaborate with researchers and annotation teams to iterate on data as models evolve. The role focuses on SFT, RL data, and synthetic/human-annotated data; remote-friendly with offices worldwide.

Qualifications

• Strong software engineering skills.
• Experience designing and building large-scale data pipelines for ML, ideally for LLM post-training.
• Track record of running distributed data processing and model inference at scale.
• Good understanding of post-training methods (SFT, RLHF, DPO) and how data shapes model behaviour.
• Excellent communication skills.

Responsibilities

• Build and maintain data pipelines for SFT, preference, and RL post-training data.
• Own dataset quality, versioning, and reproducibility.
• Turn research data recipes into production-grade, reusable components.
• Scale model inference for data generation and model-based grading.
• Design data formats and schemas to keep datasets consistent and efficient to train on.
• Collaborate with researchers and engineers to keep data pipelines aligned with model needs.

Skills

Software engineering

Designing data pipelines

Distributed data processing

Post-training methods (SFT RLHF DPO)

Communication

Job description

Cohere is a leading security-first enterprise AI company building foundation AI models and end-to-end products. We are hiring for a role that builds and owns pipelines for post-training data used to train large language models.

You will collaborate with researchers and annotation teams to iterate on data as models evolve. The role focuses on SFT, RL data, and synthetic/human-annotated data; remote-friendly with offices worldwide.